3 4 4 5 b 0351219 B
k
ORNL/IM-l1!Z7
Chemical Technology Division
SUMMARY REPORT ON TICIE DEMONSTRATION OF THE D m X K PROCESS FOR TREATMENT OF MIXED-WASTE CONTAMINATED GROWWATER
Suman P. N. Singh Thomas F. Lomenick
Date Published - April 1992
Prepared for 1J.S. Department of Energy
Office of Technology Development Budget Activity No.
E W40804OO
Prepared by the OAK R I D G E NATIONAL LABORATORY
Oak Ridge, Tennessee 37831-6285 managed by
MARTIN MARIETTA ENERGY SYSTEMS, INC., for the
U.S. DEPARTMENT OF ENERGY under contract DE-AC05-840R2 1400
~ . , ~ ~ ~ ~ M ~ , ~ ‘ ~ ......................................
FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
EXECUTIVE su MARY .......................................
ABSTRACr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IJCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ENT PROCESS .................................
4. DEMONSTRATIION DETAIL§ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. TESTRESOL,TS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. CONCLUSIONS /AID ~ ~ ~ ~ ~ M ~ , ~ ~ A T ~ ~ ~ ~ ....................
7. REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDIX A. Report by Duratek Corporation entitled, “Demonstration of Exchangc Technology for Treatment of Mkcd Waste ~ ~ ~ ~ ~ ~ ~ r ~ a t e ~ Groundwater“ . . . . . . . . . . . . . . . . . . . . . . . .
V
vii
k
xi
xv
1
3
11
17
23
35
39
41
... 111
ACKNOWLEDGEMENTS
The authors exprcss thcir apprcciation to the following Martin Marietta Energy
Systems, Inc. (Energy Systcrns) staff for their assistance with this project:
a Jennifer Kastcn €or rcscarching the data on the groundwatcrs at the Ulnitcd SLa?es
Department of Energy (DOE) sites managcd by Energy Systcrns and for developing
the treated watcr quality criteria rcferrcd t o in the report as the "Energy Systems
Treatcd Groundwatcr Quality Criteria";
Thc bid cvaluation team mcmbers who, aside from the authors, consisted c>€ Jack
Elrod, Kod Kimmitt, and Jim Wilson;
Charlcne Patrick for typing the report and Kathryn KingJones for technical cditing
of the manuscript; and
Other Energy Systems staff who revicwcd the draft reports and provided critical and
useful comments.
0
0
0
V
12 1. Block flow diagram For treating mixed-waste contaminated water . . . . . . . . .
2. Block diagram of systems for stripping waste-loaded A and C columns . . . . . 13
3. Slcctch of the bench-scale trcalrncnt sclacme to trcnt the surrogate 18 mixed-waste contaminated ~ r ~ ) ~ n ~ ~ a ~ ~ ~ ............................
4. Process flow diagram of the Duratek pilot plant to treat the 20 rnixcd-wastc contarnina ted water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Layout of the Duratek pilot plant to treat the rnkcd-waste 22 contaminated watcr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
...
vii
. TABLES
1 . Proposcd constitucnts for surrogate groundwater and desired treatment standards ...............................
2 . Energy Systcms-adoptcd treated groundwatcr quality criteria . . . . . . . . . . .
3 . Estimated trcatment capacity of thc DurasiP ion-exchange media . . . . . . . .
4 . Required contaminant concentralions in the surrogate mixed-waste groundwater and the target decontamination factors to be achicved by the treatment process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5 . Test results after passing influent through air stripper and column 1A . . . . .
6 . Test results after passing influent through column 1A . . . . . . . . . . . . . . . . .
7 . Test results after passing influent through column 1C . . . . . . . . . . . . . . . . .
8 .
9 .
Tcst results after passing influcnt through column 2D
system during first week of operation
. . . . . . . . . . . . . . . . .
Analytical data from integrated benchscale water decontamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 . Pilot plant operating data after 2 wccks o f operation . . . . . . . . . . . . . . . . .
11 . Analytical data from the pilot plant after 2 wccks of operation . . . . . . . . . .
4
6
15
24
25
26
27
28
29
31
34
ix
This projcch consists of dernonst rating an innovative process from the private sector
for the treatment of mixed-waste contaminated groundwater (that is, groundwalcr
contaminated with I adioactive and humrdous/taxic compounds) found at f he Department o f
Energy (DOE) industrial sites managcd by Martin Marietta Energy Systems, Tnc. (Energy
Systems). ‘b’hese sites include the Oak Ridge Reservation (Oak Ridge National Laboratory,
thc K-2s Site, and the Y-12 Plant) in Oak Ridge, Tennessee; the Paducah Gascous Diffusion
P1:nnt in Paducah, Kentucky; and the Portsmouth Gaseous Diffusion Plant in Piketon, Ohio.
Groundwaters at thcsc sites have bccn found to bc contaminated with radioactivc species
(chict’iy uranium238 and tcchnet ium-99) and organic and inorganic bamrdous compounds,
such as p ~ ~ ~ y ~ h l o r i n a t e d biphenyls (PCBs), h e n x n e , trichlorocthane, and barium, cadmium,
and chromium salts. ‘rho objective of this pro.ject was to identify and demonstrate an
innovat ivc process that could be used to satisfactorily treat the mixed-waste cnntamindted
water to mcet drinking watcr quality standards.
Following the prescribed evaluatcd procurement guidclincs, the proposal submilkel
by a team composed o f Duratck Corporation (Iluratek) arid the Vitrcous State Laboratory
(VSL) of tlic Catholic LJnivcrsity of America located in Washington, D. C. was sclected for
demonstration. Their proposal consistcd of dcmonstrating the Duratek process for trcating
the mixed-wastc contaminated groundwater. Briefly, this proccss consists o f treating the
contaminated watcr by air sparging and thcn passing it through a series of columns containing
Duratck proprietary Durasi10 ion-exchange media. The ion-exchangc mcdia rcmovc thc
radioxtive and hazardous compounds from thc watcr to levels below national drinking watcr
quality standards. Duratck and VSL Jcvcloped the proccss and demonstrated it in a pilot
plant capable of processing 1 gilimin o f thc contaminated watcr. T h e treated water from the
process can bc used for othcr operations at thc site.
Following breakthrough. the spent ion-exchange media loaded with the contaminants
arc removed from the water treatment circuit for regeneration of thc media and recovery of
the contaminants. During thc regeneration operation, the radioactive contaminants arc
scparatcd from the h;i~ardous species using anothcr proprictary DurasiP ion-exchange
medium.
'I'he secondary wastes from the process will likely consist of spent activatcd carbon
traps (containing the volatile organic compounds), spent chiarcoal-based media (containing the
radioactivt: spccics), and spent glass-based mcdia (containing the hamrdous components).
T h e carbon-bascd spent media can bc incinerated and the spent glass-based rncdia can be
vitrificd resulting in further rcduction o f thc secondary wastes from the process and
encapsulation o f the haardous compounds. In addition, if 21 reverse osmosis unit is used to
rcduce the coilcentration o f thc nontoxic species in the watcr (such as nitrates and sulfates
of alkali mctals), a retcntatc streem will be gcncratcd that would rcquire disposal. T h e
disposal of the retcntale dcpcnds upon its salt concentration. For example, it can be
discharged t o the wastewater treatmcnt plant, biotrcatcd, o r converted into a salt cake for
disposal in a landfill.
The pilot-plant dcmonstrations showcd that the Phrasil@ ion-cxchangc media have
high capacities for removing the contaminants from the watcr thereby proving superior to
convcniional polymeric ion-exchange rcsins. Prcliminary cost cstimatcs by Duratck
xii
Corporation for a full-scalc treatment facility places the treatment costs at about lO@/gal of
mixcd-waste contaminated water treated. Thcsc estimates do not include analytical costs or
the costs for the disposal of the sccondary wastes from the process. The costs associated with
this treatment facility are somewhat higher than the cost of similar watcr treatment using tIic
conventional ion-exchangc resins. However, the superior pcrformance and high capacities of
the DurasiP media and thcir ability to separate the radioactive contaminants from the
hazardous species should mitigate the higher estimated treatment costs.
The details of the demonstration as well as conclusions and recommcndations on the
Duratek proccss are givcn in this rcport. The major conclusion is that the Duratek process
can effectively removc the radioactive and hhc hazardous waste species in the mixed-waste
contaminatcd water to below currcnt and proposed drinking water quality standards. The
major recommendation, bascd on the pilot-scale tests, is that the Duratck proccss should be
considered for rcmcdiating thc mixed-waste contaminated groundwater found at the Encrgy
Systems-managed DOE sites. Further dcmonstration and testing of the Duratek process
should enhance its applicability and help in determining thc actual costs of treating mixed-
waste contaminated waters.
xiii
....
Sunran 19. N. Singh ThcPnaas F. hnnenick
T h i s report presents the results 01 the desnonstrationi caf thc I3uratck pc;ccss for
removal of radioactivc and h a m d o u s waste cornpounds from mixed-shinste coiitamiaia?.u.i
graundwaters found at the Department of Er1crgy (DOE) sites man:igcd by MdrrlidB M3arictl;a
Energy Systems (Energy Syslerns). The process uscs Duratek ~~~~~~~e~~~~ Dwr asif@ ion-
exchange media to remove thc above contaminants f rcm the water to podt i c ~ , t ~ ~ a t c d wntcr
that can mect current and proposed drinking water quality staradaids with r q p r d to thc above
contaminants. The demonstration showed that hhc process is simplc, colapact, versatile, and
rugged and rcquircs only minimal operator attention It is thus ~~~~1~~~~~~~~~ that this
process bc considcred for remediating Ihe mixed-waste contaminated waters foura
Energy Systems-managed DOE sites.
......
1. INTRODUCllON
. .
This project was undertaken to assist the U.S. Department of Energy (DOE)
identiljring treatment methods that could be used to treat mixed-waste contaminated
groundwaters ( i c , groundwater contaminated with radioactive, toxic, and/or hazardous
compounds) found at DOE industrial sites. Due to past industrial practices, tkc groundwater
at many DOE sites has become contaminated with radioactive and thasardoudtoxic
compounds. Thc satisfactory remediation of these contaminated groundwaters is of
considerable concern t o DOE as pressure from regulatory groups and envicsiirncntal agencies
mounts. While the bulk of the contaminants can be removed from the water using available
commercial processcs, there is a dearth of processes that can be used to remove the residual
contaminants down to the low levels currently required by regulations.
The objective of this project was to demonstrate the efficacy of innovative privatc-
scctor devcloped treatment processes that could be used to satisfactorily remove radioactive
and hazardous constituents down to the desircd levels from contaminated groundwater on
DOE, Oak Ridge Field Office (DOE-OR) sites operated by Martin Marietta Energy Systems,
Inc. (Energy Systems). Current statutes, such as thc Comprehensive Environmental
Response, Compensation and Liability Act (CERCLA) and the Resource Conservation and
Recovery Act (RCRA) and their amendments, have set strict guidelines for the remediation
and protection of the nation’s groundwater. These regulations guide the groundwater
1
2
monitoring and protection plans at the Energy Systems-managed facilities. Therefore,
remedial action plans and appropriate treatment processcs are needed to treat the
contaminated groundwater.
The goal of this projcct was not only to identify but also demonstrate treatment
operations thdt could succcssf~~lly dccontainiiiate mixed waste contaminated groundwater to
meet Energy Systems’ adopted treated groutidwater quality criteria. If sucln technology was
deinonstratcd, then it could be successfully employed to treat the water not oiily at the DOE
sites managed by Encrgy Systems but at other similarly contaminated sites nationwide as well.
‘]The scope of the project consisted 01 the following stcps:
38 Dcvcloping the composition of a candidate water that reflects thc composition of the
~ ~ ~ ~ ~ ~ a i ~ i ~ ~ a ~ c ~ groundwatcrs fourmd sat thc five DOE sites managed by Energy Sys terns
(hereaftcr rcferrccf to as Ener Systems-rn an aged sites. 1 The c ~ ~ ~ c ~ ~ ~ ~ a t ~ ~ ~ ~ ranges
of thc hazardous ~ o ~ p ~ ~ ~ ~ ~ ~ s known to occur in the g r o ~ ~ ~ ~ a ~ ~ ~ s at the Ener
Systems-managcd sites are given in Table 1. Table Z was devcloped based on
~~~~~~~a~~~~ obtaincd f m m sihe e n v i ~ ~ ~ n i ~ e r ~ ~ a l ~ ~ Q Q I ~ S by Rogers, et aP. ’-’ This
information was reviewed along with inore current analytical d a h on some
g ~ o ~ ~ ~ w a ~ ~ ~ ~ at the Y-12 plant provided by ~ ~ ~ ~ o ~ ~ ~ 4 to arrive at the list given in
Table 1.
Q ~ ~ ~ a ~ ~ ~ ~ ~ ~ n ~ thc desired lcvel to which the hazardous compounds must bc removed
from the surrogate water by the treatment process. The treatment goals f o r the
contaminants are also given in Table 1. These goals weie obtained from thc Energy
Systems adopted treated groundwater quality criteria that are shown in Table 2.
These valucs werc devcloped bascd on the premisc that the treated groundwater
should bc ablc to meet the U.S. Environmental Brotcction Agency’s current and
proposed regulations for the contaminants listed in Table 2 in drinking water. The
information for devcloping the criteria given in Table 2 was obtained from the Code
of Fedcral ReguIations’ and thc Federal R ~ g i s t e r . ~ , ~
3
4
Proposed Desired treatment Cornpound concentrations (mg/L) standards (mg/L) ~--
Barium
Cadiniiim
Chromium
Carbon tetrachloride
'I'c t r acblo roet hylene
'rrichiorQeti1ane
Vinyl chloride
Methylene chloride
Benzene
Toluene
Xylenes
10-100
0.1-1.0
0.1-1.0
10-100
0.01-5
0.01-5
0.01-5
0.01 -2
0.01-2
0.01-5
20- 1 0 0
50-100
Nitrates
30-250
Radionuclides
0.1-1"
10,008-1ooob
1.0
0.01
0.05
1 .o
0.005
0.085
0.005
0.002
0.002
0.005
2
10
10
0.01
900"
PCBS 1 .001
5
- Tabk 1 (continued)
Proposed Desired treatment Compound concentrations (m&) standards ( m a )
Calcium
Chloride
Iron
Magnesium
Manganese
Potassium
Sodium
Zinc
Sulfate
Conductivity
50
1
2
30
.01
2
5
10
30
300 pmhosfcm
mDd
25
0.3
TBD
4.5
5.0
25
TBD
... “Present as U,O, bPresent as pCiL of TCO,. ‘pCi/L. dTBD = to be determined.
ArSCIliC
Barium
Cadmium
Chromium
C o P P
Fluoride
Iron
IR2d
Mangancse
Mercury
Nickcl
Nitrateb
Selenium
Silver
Sulfate
Thallium
Zinc
pH"
Total suspcndcd solids
Total dissolved solids
Oil and grease
Total residual chlorine
Total toxic organics
PCBd
PC"
Gross a lphd
Coinbincd radium226 and radium-228
0.05
1 .0
0.01
0.05
1 .o 2.4
0.3
0.05
0.5
0.002
0.1
10.0
0.01
0.05
408
o.oO05
5
6.519.5
31.0
500.0
26.0
0.1
2.13
0.001
o.Ooo5
15 p@ifL
5 pCi/L
7
Parameter
Grass beta 4 mremfyear
Radium-223 1 p a / %
Strontiurn-
Technetium99
orium-23 1
Thorium-234
Tritium
Zirconium95
Benzene
Carbon tetrachloride
1,l-dichlorocthylene
1,2-dichloropropane
Tetrachlormthylene 0.005
To 1 uene 2
1,l ,l-trichloroethane 0.2
Trichloroethylene 0.005
Trihalomcthanes (totaly 0.1
Viiiyle chloride 0.002
Xylcnes 18
“All values are in mgiL unless indicated otherwise. bIndicates nitrate present as nitrogen. ‘Dimensionlcss. dIncludes Aroclor 1254 and 1260. ‘Includes Aroclor 1016, 1221, 1232, 1242, and 1248. ’Includes radium-226, but excludes radon and uranium. gSum concentration of chloroform, bromoform, bromodichloromethane, and
dibromochloromerhane.
8
@ Pulsing the private, sector through a competitive bid process to idcntify suitable
(preferably innovative) technologies to treat the surrogate contaminated water by
issuing a rcqucst for proposals (RFP).
@ Evaluating industrial responses to the RFP and selecting the most promising
treatment method by using an evaluated procurement strategy.
@ Awarding the contract to the most promising process and having the private company
demonstrate the capabilities o f its process to treat the surrogate groundwater a t a
sustained rate of 1 ga lh in .
8 Monitoring and evaluating the demonstration test results, assessing the efficacy of the
process, and determining its applicability to remediate mixed waste contaminated
water.
8 Preparing an assessment report and promoting the use of the demonstrated process
to treat contaminated waters at DOE sites.
As part of the procurement strategy to obtain the desired treatment proccss an
"Expression of Interest" letter was sent to 230 prospective private companies to determine
their interest in developing/providing the technology. Scvcnteen companies responded that
they would be intcrestcd in receiving the RFP for the project. Of thcsc, four companics
responded with proposals to dcvelop/provide the decontamination proccss.
9
-. .... . .
‘ The four proposals werc comprehensivcly evaluated and ranked in accordance with
the evaluation procurement guidelines by an tcam of Energy Systems staff. Thcse guidelincs
stipulate that the vendor proposals be evaluated using a weighted point rating system. This
method objcctively evaluates thc proposals based on the following main critcria: 1) the
technical merits of the proposed approach and process to treat the groundwater; 2) the
vendor’s corporatc and tcchnical statYs expcrience in treating similar wastes; and 3 ) the cost
for performing the technology demonstration as given in statement or work. The evaluation
team consisted of five voting members. Four of thesc were engineers with cxpertisc in
radioactive waste management, groundwater treatment, and process enginccring. The fifth
membcr was a procuremcnt specialist. The bid evaluation team was supplerncnted by five
nonvoting Energy Systcms staff with considcrablc expcrience in water treatment and
environmental aft‘airs who served as consultants to the evaluation tcam.
The four proposals wcrc ranked and, based on the available funds, a contract was let
to the highest ranking proposal to undertake the decontamination project. This proposal was
submittcd by a team consisting of Duratek Corporation (Duratck) and thc Vitreous State
Laboratory (VSL) of the Catholic University of America (CUA). Duratek started work on
the project in February 1991 following cornplction of contract formalities and relcase oF the
necessary funds by DOE’S Office of Technology Development (OTD).
....
The trcatrnent process developed by Duratck (and VSL) to treat the mixed-waste
contaminated watcr essentially consistcd of removing the volatile organic compounds (VQCs)
from the water by air sparging and then passing the water through a serics of columns
containing proprietary DurasiP ion-exchange media designed t o remove the radioactive and
toxic components fr'rom the water. Figiirc I is a block flow diagram of the proccss that
illustrates the contaminants reniovcd by the various columns. Thc treated watcr from the
process can either be discharged o r used within the industrial ftt'acility whcrc the ~ ~ ~ ~ i ~ ~ ~ a t e ~
was withdrawn. The DurasilQ mcciia in columns A and C are charcoal based while that in
column D is glass hascd. The ion-exchange mcdia arc Duratek proprietary media that are
chcmically treated to trap thc indicated contaminants. These XluaasilQ media differ from
mort conventional ion-exchangc resins in that they are not polynicric resins but a rc chemically
trcatcd charcoal and glass-basccl materials that act more likc molecular sievcs. IIoward'
indicated that these Duratck ion-exchangers have significantly highcr capacitics for the
contaminants than convcntional ion-cxchangc resins.
Figure 2 is a block flow diagram for stripping thc contaminants loaded on the
DurasiP media in columns A and C (shown in Fig. 1) SO as to scparatc thc radioactive
components from thc othcr contaminants. This separation is achicved by using another
specially treated, charcoal-based DurasilQ medium loaded in column PI. This medium is
specifically dcsigncd to t rap only the radioaclivc contaminants from the rcgeneration of thc
spent columns in the primary water trcatmcnt operation. Whcn loaded, thc spcnt mcdiurn
in column H can bc removed and incinerated. The residue from incincraiion can be mixsd
11
91 A-579R
WATER AIR
CONDiTIONING SPARGER DURASIL" ,ON-EXCHANGE COLUMNS
vocs v o c s U CO N-A M ! N Ah; D C B S C r SpEC,ES REkOVED J 'ROM ThE WATEF! i C a y TEE ION-EXCHANGE C r MEDIA
L
h)
(Na+, CO': Mg+, NO:, So i )
TREATEJ WATER
-- R E V E R S E OSMOSIS
UNIT (IF REQUiRED)
Fig. 1. Block flow diagram for treating mixed-waste contaminated water
HN03
O R N L DVJG 92A-252
c w
Fig. 2. Block diagram of systems for stripping waste-loaded A and C columns.
14
with the 5pcnt glass-based medium from column D and vitrified to encapsulate the
contaminants. Table 3 givcs the estimated treatment capacities of the various DurasiP rncdia
cicvelopcd in this project.
The proccss can not only remove the mixed-waste contaminants from the water but
can also separate the radioactive contaminants from the hazardous/toxic components. This
greatly facilitates the disposal of the sccondary wastes from the process. The secondary
wastes likely to be gencrated in the process consist of spent media, spent activated carbon
(coirtaining the VOCs), and, if required, spent reverse O S I T ~ Q S ~ S media (Le., fouled
mcmbrancs), and a retentate stream containing the Na, Ca, Mg, NO,, and SO, ions. Duratek’
estimates that processing onc million gallons of the mixed-waste coinlaminated water to meet
thc desired treatment levels will result in the productiori of 25 ft3 of charcoal-based and 57
f t3 of glass-based spcnt iori-exchange media loaded with the toxic conipounds and
approximately 33 ft3 of charcoal-based spent mcdia loaded with the radioactivc components.
I11 dnlition, Duratek’ indicates that the sccondary waste volurnc can be further reduced by
at least a factor o f 10 by incinerating the charcoal-based spent media and vitrifying the glass-
based spcnt media. Further, the residue from the incincration of the charcoal-based media
could also be vitrified with thc spent glass-based media thereby encapsulating and removing
the contaminants t‘rorii the environment.
Duratek’ Corporation estimated that a facility designed to treat 10,OOO gal/d of the
mixed-waste contaminated water to the iridicated treatment standards would cost (in 1991
1J.S. dollars) approximately 10C/gal o f water treated. This cost estimate includes an allowance
for the capital cost and the operating costs for the facility but docs not include analytical costs
and the cost for the disposal of the secondary wastes from the process.
15
16
Further details on the process, the bench-scale testing, and the dcnonstratioii are
given in the followirig sections and in the report prepared by Duratek Corporation given in
Appendix A.
- .... 4. DEMON!XRATION DETAILS
The decontamination of the mixed-waste surrogatc water was acconiplished in two
phases as described below.
Phase 1: Phasc 1 of thc demonstration consisted of conducting bench-scale tests.
These tcsts were performed to achieve the following:
* Identification and/or dcvelopment of the inn-exchange media that would result in
achieving the desired removal of the contaminants from the surrogate groundwater;
Development of a treatment scheme for the surrogatc contaminatcd watcr;
Detcrmination of the significant process variables and the cptirnan
conditions for achieving the treatment goal;
Determination of the sccondary wastes likely t c be generated, their characteristics,
and disposal options; and
Identification of any problems with decontaminating the mixed-waste watcr using the
proposed treatment method.
T h e bench-scale testing was performed by VSL and basically consisted of treating the
surrogate water through small columns (6 mL by volume) of several different ion-exchange
materials. These columns contained the tailored, Duratck proprietary, ion-exchange media
0
a
*
a
designed to separate and rcmove the contaminants from the mixed-waste water. Figure 3 is
a sketch of the bench-scalc trcatment schcme developed by VSL to treat the mixed-waste
surrogate groundwater.
The results of the Phase 1 tests indicated that VSL (and Duratek) were able to not
only treat thc mixed-waste surrogate groundwater to meet the Encrgy Systems-adopted
17
1 i
'i A1 EFFLUEHl i I
WATER RAD OHCICUDES COiuTAIYi4iG
B Y 3 METAL SPECiES
Pn - METERING PLJMPS Sn - SAMPLING POINTS 1An - 6 mL D701 !Cn - 3 mL/3 rnL D701/D707G 4th - 6 mL D190 G J L - cAs/LlauiD
1An. 7Cn. ~ b n - ION-EXCHANGE COLIIkNS
f': - AS NECESSARY
Fig. 3. Sketch of the bench-scale treatment scheme 10 treat the surrogate mixed-wastc conLarnina!cd groundwater.
19
treated ~ ~ o u ~ d w a t ~ ~ quality standards, but wcrc also able to separate the contaminants in the
mhcd-waste water into two sectonday waste strearms-one containing only the radioactivc
spccics and the othcr the ~ ~ ~ a r d ~ ~ ~ s / t o x ~ c species. This scparation grcafly facilitates the
disposal of the secondary wastes because the wastes do not have to be disposed as radioactive
hazardous wastes.
ne herach-scale test rcsults are given in the next section and the dctails of the tcsting
are given in Appendix st.
.aw 2: Yhasc 2 consisted of demonstrating thc process at a pilot plant scalc
designcd to treat thc suirogate mixcd-wastc: contaminated groundwater at the rate of 1 gal/
min. “l%e pilot plant tests were conducted to achieve the follr)wing:
0 Dcmcanst rat ion of the s ~ ~ ~ a b ~ l ~ ~ y and capabilitics of the proccss by ~ ~ ~ ~ o ~ . ~ ~ ~ ~ the
treatment ira industrial-scale equipment;
~ ~ ~ ~ ~ ~ ~ ~ c a t ~ o ~ of any p r ~ ~ ~ ~ l e ~ ~ i s with operating the process at ail industrial scale such
as flooding, channeling, arid excessive pressure drop through the columns; and
Generation of data for the industrial application oE the d c ~ o n ~ ~ ~ i n a t ~ o ~ process.
‘ii’lne pilot plant operations were perfcrrmcd by Duratek and VSL at the VSL facility,
Figure 4 is the prvccas flow diagram of the pilot plant and Fig. 5 is the layout of the pilot
plant at tlic VSL facility. The pilot plant basically replicates the bcnch-scale proms on a
larger scale. The pilot plant consists of treating the mixed-waste contaminated surrogate
groundwater through cylindrical ion-exchange columns that are 6-in. ID by 5-ft long. The
volume of each column is approximately 1 €e3 (7.5 gal or 28 E). 111e first three columns of
the pilot plant arc Eabricatcd OT 304 stainless stecl (to resist attack by organics) while the
s
s,
20
-.+$ ,p,
........................ +-------.
..... +--:
;P’ ........................ np???*...-;
.. lp
I-UTp-.-- -
......
:i
.I p ”
%-:
........
d 4
..... ...
* --
-__L = -I_
+.I
m 3
c)
m 2 c” k
d 3 0
G:
NOTE: ACCESS INTO TANK SHALL
BE PROVIDED
FIXED WORK I / EXISTING BENCH SINK
FIXED WORK BENCH
I I L- ._ INFLUENT
1 , r ~ I X DRUM GAS PURGING / AT 100 gal C 0 LU k N -\ ! !
-
I 0 Z iil m Y LY 0 3 n
x W
G
12-ft SQUARE x 1.5-ff DEEP
/--- CONTAINMENT W!TH LIQUID
DETECTION MONITOR (1 5.000-gal EMPTY VOLUME)
FIXED WORK BENCH
PILOT PLANT ROOM INTERIOR Fig. 5. Layout of the Duratek pilot plant to treat the mixed-waste contaminated water.
22
rcmaining columns are made from PVC or clear plastic pipe. Coliirnns 1'4, IC, and 2D each
contain 6 gal of ion-exchange media.
Further details of the pilot plant operation are given in Appcndix A, and the test
rcsidts are given in the next section.
c
...
...
e results obtained from the bench-scale and the pilot lane treatment of thc
surrogate mixed-waste contaminated groundwater using Duratek's ion-exchange process are
summarized in this section. Additional details and analysis of these rcsulhs are given in
Appendix A.
Thc reyuircd conccntsahms of the ~ o ~ ~ ~ ~ ~ ~ n a ~ ~ ~ ~ iia the surrogate mked-wastc watcr
beforc and after treatment and the respccbive target decontamination Factors (DFs) arc given
in Table 4. It should be noted that, because of the solubility limits for certain ions, the pH
uent water to the treatnzent train was adjusted to 5.5 to prevent mine of the toxic
elements from precipitating out of the surrogate groundwater. When trcating actual
groundwaters, such conditioning may or n a y not bc required depcnding upon the quality of"
the groundwater and the treatment objective. The bench-scale test results arc given in Tables
5 through 8 and summarized in Table 9 for thc integrated bench-scale treatment unit. The
test results obtained from the operation of the pilot piant arc: summarized in Table 10.
An ion-cxchange column is designed to remove thc contaminants from the fluid by
trapping them on the mcdia as thc fluid passes through the column. Its effectiveness can be
measured by two factors:: 1) the concentralion of the contaminant in the eClluent from the
column, and 2) the volome of inlluent that i s treated by the column before "breakthrough"
of the contaminant in the treatcd fluid exiting the column. Because these Iwo factors arc
significant in evaluating the tcst results, a brief description of these two factors is given before
discussing thc tcst results.
23
24
Required concentrations ( m a ) Target deconta- mination factor
(DF) Contaminant species Before treat ment After treat ment
Barium
Cadmium
Calcium
Chromium
@o@lper
Iron
Magnesium
Manganese
Potassium
Sodium
Zinc
Technetium-99
Uranium-238
Chloride
Nitrate
Sulfate
Carbon tetrachloride (CCI,)
Tetrachloroethylene (C;Cl,)
Tr icb lo roet ha ne (CH,CCI,)
Vinyl chloridc (CH,CMCB)
Methylene chloride (CM,CI,)
Benzene (C,H,)
Tolucnc (C,H,CH,)
Xylene [C,H,(CH,),]
PCB
10- 100
0.1-1
50
0.1-1
10-100
2
30
0.01
2
5
10
5.9 x 10-~ to 5.9 x io-, 0.1-1
1
30-2.50
30
0.01-5
0.01-5
0.01-5
0.01-2
0.01-2
0.01-5
20.- 1 00
5&100
1
1.0
0.01
TBD”
0.05
1.0
0.3
1BD
0.5
TBD
4.5
5
5.3 x 10-5
0.0 1
25
10
25
0.005
0.005
0.005
0.002
0.002
0.005
2
10
0.001
100
100
TBD
20
100
7
TBD
0.02
TBD
1.1
2
11
100
0.04
25
1.2
1000
1000
1000
1000
1000
1000
50
10
1000 ~ l...l.l..l..
“’I’UD = to be determined.
25
Table 5. Test results after passing influent through air stripper and column 1A
Conc in Conc in Conc in effluent effluent influent" of air DFair of Col DF DF
Species (ppm) strippep stripper 1A" 1A Total
(PPb) (PPb)
Carbon tetrachloride 5 <1 >5,000 <1 >5,000
Tetrachloroethylene' <1 <1 <1
Trichloroe t hane 75 1,855 40 6 309 12,500
Mcthylcnc chloride"
Bcnzene 75 65 1,150 <1 >65 >75,000
Toluenc 215 300 720 41 >300 >215,000
Xylenes 70 40 1,750 <1 >40 >70,000
"Flow rate of influent (FtnE) = 0.53 cm3/min. bFlow rate of air (FaJ = 100 cm3/min, FJF,nf = 189, and height of influent bubbled = 13 cm. cCOlumn volume = 6.0 cm3 and residence time = 11.3 min. 'Tetrachloroethylene and vinyl chloride (a gas) were found to he too volatile to remain in
Tenative analytical data for methylene chloride suggests DF values in the air stripper of solution.
around three orders of magnitude.
7.6
c u
Cr
Cd
IJ
1 ‘c
Mg
Ca
Na
K
c1 NO,
so, PCB
1-70
10
6
lo0
1100
50
1 0.25 4%Ml m 0.001 2 <0.025 > 48 >508
30 NA“
5n N A
6.5 NA
3.5 N A
N A
”4
Nf?
N A
N A
N A
N A
N A
1s N A N A N A
228 NA N A N A
30 N A N A N A
4 < 1 > 4 m >580
“Colurnn paiamcters: flow ratc of influent = 0.5 cm’/min, Cfolumn volume = 3 crd, and
bNumber of column volume\ of influcnt p a w d through column before DF falls below
‘NA =: Not applicable. T h e column is virtually transparent t o these ions during the run.
rcsiticncc time = 6 nrin.
100 for Cu, CX, and I J; 40 for Tc; and 4OOO for PCB.
27
U 1 41.3 3 7
Tc $.MI2 < 0.025 > 48 > S
...
30 NA"
Ca 50 NA
Na 6.5 NA
K 3.6 NA
Cl 15 NA
NO, * 228 NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
so4 30
PCB 4
NA NA NA
<1 r4aM >5M
"Column parameters: flow rate of influent = 0.5 cm3/min, a lumxi voiollume = 3 ern3, and
bNurnber of column volumes of inkluent passed through colunin before DF falls below 100
'NA = Not applicable. The column is virtually transparent to tbese ions during the run.
residence time 2= fi min.
for Cu and U; 40 for Tc, 20 for Cr and 4000 for PCB.
28
Sa 10 0.2 50,000 $00
c3d 1 <0.025 > 40,
CU 10 6.7 1500 > 1200
Z n 10 4.0 2500 800
Fc 2 30
Mg 30 NA’
Ca 50 NA
Na 7 NA
M 2 NA
c1 15 NA
45 >m N A NA
NA
NA
NA
NA
NA
NA
NA
NA
NO, 360 NA NA NA
“Column parameters: flow rate of influent = 1 cm3/min, column volume = 6 cm2,
bNumber of column volumes of influent passed through column before DF falls below
‘NA = Not agpliable. The column is virtually transparent to these ions during the run.
residence time = 6 m h
1 0 for Ba, GI9 Cu, and Zn; and before DF falls below 10 for Fe.
29
Table 9. Analytical data from integrated tpeachscaIe water ~~~~~~~~ system during first week of operation
Species Concentration at sampling points" @a) Air sparging CQiUnm and column 1A
s1 s2 3
... .. .
Technetium-99
Uranium238
Chromium
Copper
Iron
Cadmium
Calcium
Magnesium
Sodium
CCl,
CLCb
CH3CCI,
CH,CHCI
CH,CI,
c6&
C6WH3
C,WCH3)2
PCB
0.85
1,163
1,109
13,553
3,780
1,014
75,000
39,000
11,ooo
345
730
1,720
b
h
1,980
82,500
57,500
700
b
b
b
b
b
b
b
b
b
10.6
18.2
323.5
b
h
22.2
820
1,465
column 1c
0.08
0
224
0
79
-183
0.0s
0.1
0. I
b
b
0.02
1.55
0.6
0.4
Uranium-238
Chromium
Copper
Iron
Cadmium
s4 s 5 s6
0
3%
5,433
147
1,067
0
7
29
115
1,076
0
4
0
105
1,095
30
copper 5 13 8
Iron 84 44 72
1 CAdmiurn 580 _I
1 - "See Fig. 3 for sampling point locations. bDaba not provided
....
32
The concentration of the contarninant in the effluent from an ion-exchange column
is a measure of the effectivencss of the media to trap (remove) the contaminant from the
influent. The lower the concentration of the contaminant in the effluent (compared to its
concentration in the influent) the more effective the column in treating the contaminated
fluid.
The capacity of an ion-exchange column, on the other hand, is the measure of the
volume of contaminated fluid that can be treated by the column before the media in the
column becomes saturated with the contaminant, and the contarninant essentially flows
through the column without being trapped. At this point, the contaniinant is said to
"breakthrough" the bed. The capacity of an ion-exchange column, therefore, is the volume
oE influent that is treated by the bed before it experiences breakthrough. This volume is
often measured and reported in terms of the number of "column volumes" o r CVs. The CV
is the volume of fluid equal to the volume of the ion-exchange bed in the column. Therefore,
the larger the number of column volumes processed by an ion-exchange column, the more
cffective the media in treating the contaminated fluid.
Most ion-exchange columns are arranged so that the fluid passes through them in
series. When breakthrough occurs in the first column in the series, the contaminants are
trapped in the second column. The first column is thcn removed from service, the second
column becomes thc lead column, and a fresh or regenerated column is added in series after
the second column and the treatment operation is continued. The medium in the first column
can either be regenerated or replaced with fresh medium and the column can be returned to
service to continuc treating the contaminated fluid.
33
With the above background, it can be seen from reviewing the results given in Tables
5 through 8 that the DurasiP ion-exchange media can readily remove the indicated
contaminants from the surrogate mixed-waste contaminated water to below the required
levels. In addition, these media appear to havc high capacities for treating the contaminated
water before experiencing breakthrough. Thc data in Tables 10 and 11 show that similar
effective treatment results were also obtained in the pilot plant opcratiom. Therefore, it
appears that the Duratek process should be able to satisfactorily treat mixed-waste
contaminated water similar to the surrogate groundwater cornpition.
If thc concentrations of thc nontoxic ions in the treated water after the DurasiP
columns are higher than acceptable values, the effluent from thc columns can be passed
through a reverse osmosis unit to reduce the concentrations ol these ions. In addition, the
Duratek process is deliberately designed t o separate the radioactive components in the
secondary wastes from the hamrdous/toxic species by using specially tailored Durasilg media.
This separation greatly reduces thc problems associated with the disposal of mixed radioactive
and hazardous wastes.
34
of zp
- - ~ SpHi635 Concentrations at Sampling Points" @g&)
v1 V2 V6 v 7 V8 v10 v11 I -...- -
IMOW0.4aNIa
Barium
Cadrninm
c;nlclunab
Chromium
CQIPPer
Iron
Magnesiumb
Manganese
Potassiaim
S d i U R I b
Zinc
Technetium
Urali.,illam
ORGANICS
C a r h m tetrachloride
Tetrachloroethylene
Trichlsroethane
Vinyl chloride
Methylene chloride
kw7xnc
Toluene
Xylenes
PCB
47
1,081
75
662
6,260
0
58
0
c
13
0
0.4211
1,275
43 46
%7 885
70 90
576 5
5,850 6,751
0 0
47 48
0 6
c C
13 36
0 0
0.4860 0.039
1,129 40
1,210 22
1,185 74
0 0
C c
C C
332 369
43,510 1,894
31,725 2,575
c c
0
0
0
c
C
0
0.082
0.058
c
45
824
92
0
734
0
49
6
C
41
0
0
C
c
c
c
c
C
c
C
C
48
873
162
10
i,32a
239
193
9
c
6
8
3
c
c
c
c
c
C
C
C
c
0
0
181
2
0
109
202
0
C
6
0
2
c
c
c
C
c
c
c
c
c
0
0
111
1
0
82
123
0
c
6
0
3
c
c
C
C
C
c
C
c
c ~
"See Fig, 4 for sampling p i n t locations. b ~ ~ h e data for calcium, magnesium, and sodium are in units of m&. 'Data not provided.
... .. .
Based on the results of thc demonstration, the following conclusions can be drawn
regarding the Duratek ion-exchange prcmss:
B "he process can remove the radioactive and the hazardous waste corn
in the water tra helow drinking water quality kvcils.
The process is innovative in that it not only removes the ~ ~ c d - ~ a ~ t ~ ~
~ ~ n ~ a ~ ~ ~ a ~ ~ s from the water but also separates them into a radioactkc waste stream and a
hazardous washe stream, which greatly facilitates ultimate disposal.
Q The DurasiP ion-exchange media used in the process appear to have a high
capacity for removing the contaminants from the watcr.
e The pilot plant operated without any major problems and dernonstratcd that
the bench-scale data can be readily sealed up to largcr operations. Ehvever, duc to project
~ ~ ~ ~ d ~ n ~ and schedule constraints, the pilot piant operations had to he curtailed to meet the
funding and schedule limitations. At s ~ ~ t d o w n of the pilot plant operations, there was no
breakthrough of thc contaminants through the ion-exchange media. Therefore, it was not
possible to measure the ultimate removal capabilities of the ion-exchange media o r 10 more
closely estimate the life-cycle costs of the process.
8 The process generates minimal secondary wastes. These wastes consist of
spent activated carbon traps, spent charcoal-based media (containing the radioactive species)
and spent glass-based media (containing the hazardous compunds) . For example, these
wastes can Be further reduced in volume hy incinerating the carbon-based wastes and
35
vitrifying the glass-based media.
encapsulate the hazardous compounds from the environrncwt.
Vitrifying the spent glass-based media will effectively
cgo The process is simple, compact, and rugged. It requires minimal operator
attention. The process equipment can be designed to be either skid-mounted or built as a
mobile treatment unit so that, if necessary, it can be moved into the field to treat the
contaminated groundwater close to the wellhead.
a The preliminary economics of the process appear to be fair. The cost
estimates were developed based on limitcd pilot plant operations and seem to suggest that
a full-scale process may be an economical means of treating mixed-waste contaminated water.
However, it should be noted that because of liability concerns with the shipment of mixed
wastes from DOE sites, the demonstration had t o be conducted on surrogate waters tailored
to reflect the actual contaminated groundwatcrs found on all the Energy Systems-managed
DOE sitcs, Because of this restriction, the surrogate water composition was designed to
reflect the worst-case compositions of the actual groundwaters found at the sites. In all
probability, the actual contaminated groundwaters would very likely be a subset of the
surrogate water composition and the Duratek process could be designed to readily and
economically treat the actual contaminated groundwaters.
The following recommendations are made based on the above conculsions and the
nesults of the demonstration:
9 The Duratek process should be considered for the treatment of the contaminated
waters found at the Energy Systems-managed DOE sites.
To avoid the liability concerns associated with shipping mixed wastes from the DOE
sites, a mobile unit based on the Duratek process should be built that can be takcn
e
37
....
to the DOE sites and tested con the actual mntaminaled groundwaters. This testing
will help establish the capabilities and the processing costs for the technology.
...
7. REFERENCES
..... .
1.
2.
3.
4.
5.
6.
7.
8.
9.
J. G. Rogers, et al., Oak Ridge Resewation Environmental Report for 1988, VQ~WTWS I and 2, ES/ESN-8/VIL & V2, Martin Marietta Energy Systems, Inc., Oak Ridge, Tennessee, May 1989.
J. G. Rogers, T. G. Jett and M. R. Aaron, Paducah Gaseous Dqfusion Plant Site Eaviro~rnmtal Report for 1988, ES/ESH-8/V3, Martin Marietta Energy Systems, Inc., Oak Ridge, Tennessee, May 1989.
J. G. Rogers, et al., Pommouth Gaseous Diffusion Plant Site Environmental Report for 1988, ES/ESH-8/V4, Martin Marietta Energy Systems, he., Oak Ridge, 'Tennessee, May 1989.
C. W. Kimbrough, Martin Marietta Energy Systems, Inc., Oak Ridge, Tennessee, personal comrnunication to J. k. Kasten, Oak Ridge National Laboratory, Oak Ridge, Tennessee, July - August 1989.
U. S. Erivironmental Protection Agency, "National Drinking Water ~ e g ~ ~ a t ~ o ~ ~ , " 4 ~ Code of Federal Regulalions, Pts. 141-143 (July 1, 1991).
U. S. Environmental Protection Agency, "National Primary and Secondary Drinking Water Regulations," Fed. Reg. 54(97), 22062-2216Q (May 22, 1989).
U. S. Environmental Protection Agency, "National Drinking Water Advisory Council; Open Mccting," Fed. Reg. 54( 149) 321 16-321 17 (May 22, 1989).
S. J. Howard, Duratck Corporation, Columbia, Maryland, to S. P. W. Singh, Oak Ridge National Laboratory, Oak Ridge, Tennessee, March 27, 1991.
R. E. Sassoon, et a]., Demonslratim of Ion-Exchange Tec~inology for Treaiment of Mixed Waste Contaminated Groundwater, Vitreous State Laboratory of the Catholic University of America and Duratek Corporation, December 20, 1991. (Sce Appendix A).
39
APPENDIX A
Report by Duratek Corporation entitled "Demonstration of Ion-Exchange Technology for Trcatment of Mixed Waste Contaminated Groundwater"
...
...
Demonstrat ion of Ion-Exchange Technology for Ta atment o
Contaminated Graun
Final Report
arietta Energy Syste Oak Ridge, TN
December 2
Y
Richard E. Sassoon, Hamid Hojaji, Noel Batitactac, arek Brandys and Pedro B.
Vitreous State k olic Univers
43
....... Demonstration of' Ion-Exchange
Technology for Treatment of Mixed
Waste Contaminated Grsun
Final Report
December 20, 1991
BY ...
Richard E. Sassoon, Hamid Hojaji, Noel Balitactac, Marek Brandys and Pedro B. M a d o
Vitreous State Laboratory The Catholic University of America
and the Duratek Corporation
45
47
A system based on selective ion exchange was devel oval. of mixed waste
contaminants from ~ n ~ r n i ~ t e ~ ~ ~ o u n d w a ~ ~ . Selecthie ion exchangers are materials that
remove just the contaminant ions from the wastesit
concentrations such as I%+, Mg2*, Ca2’
materials therefore have much higher capacities and become more cast effective as eam
to other ion exchange resins.
while allowing ~ t o n - t ~ x i c ions at Rig
O i and SO:- ta pass through tke system. Such
A mix& waste contaminated groundwater composition ~ e ~ ~ s e ~ ~ t i v e
contaminated mixture of ground waters found at 5 DOE site
Energy Systems was test
contained uranium and ~ h ~ e ~ u ~ as radionuclides, volatile organics, PCB’s, ~ ~ ~ r ~ ~ a ~ e md ~ Q X ~ C
metal cations. The system tested was successful at removing all the ~ ~ ~ ~ i n ~ t s to below their
safe drinking water levels. The decontamination system consisted of the following stages:
by Martin Marietta
An air sparger which reduces levels of the volatile organics by over an order of
magnitude;
“A” media which remove all. the remaining organics including the BCB’s, a l l the
technetium and most of the uranium and chromate in solution;
“C” media which remove the remaining uranium and chromate from solution;
An additional “A” column which acts as a guard column to protect the last stage
of the system;
“D” media which remove all the toxic metal cations in solution.
49
with a residence time of 6 minutes for
mks 0% 1 muminu system mQ 1 gpm in thc.pilot system. @er
a d through th
som CY’S for A media, over 2
determind.
CV’S for c m d i a m
The wand pat of this study covered the a t i o n of the mix radioactive md toxic
on the A mQ C rn ia. Passage of 1N HNO, through an A or C
both the radioactive m d tsxk inorganic wmpsaents that had been capturd on
the columns. The effluent of the stripping process was then passed ahrough msther &xtive
removed just the radioactives from the solution and allowed
the toxic mmpraetits to pass through. In this way elimination of m y f ind mixed waste was
achieved a d the media may be regenerated.
areas still remain available for study in order to obtain more accurate cost
estkmatees for the system. These ~~~~d~~
(a) ~~~~~~~~~ of ~p~inti~~ of tke pilot plant to determine the true capacities of the ion
ia rather than lower iimits;
@) ~~~~~~~~~~ of the
(c) determination of the number of regenerations of ion exchange media that may
(d) operation of a demonstration wit on actual groundwater,
ing process at the pilot mle level;
50
....
Benchde System
e sized unit proposed and used in d m n It consists of an en^ n
nerets of the influent are mixed with the organic corn es sf air ~ w ~ ~ l ~ ~ g amd ion exchange columns. Peri
in Figure 2,
s in the unit to ensure a constant Row rate of solution through the system.
51
Table 1: Surrogate Water Corn
equired Concentrations (nmgfi) DF
--.-_ Before Tratmeni After ~~~t~~~~
10- 1 1 .o 1W Cadmium a1-1 0.01 100
calcium 58 TBD TBD
Chromium 0.1-1 0.05 20
18-1 1 .0 1
Iron 2 0.3 7
Mag n s i u rn 30 TBD TBD
0.01 0.5 0.02
2 TEdD TBD
Sodium 5 4.5 1.1 Zinc 10 5 2
TWh~&Um-99 59 O"ooo053 11
0.1-1 0.01 100
52
....
Table I (continued)
...
Species
Chloride
Nitrate
Sulfate
Carbon tetrachlo- ride
Tetrachloro- ethylene
Trichloroethane
Vinyl Chloride
Methylene Chloride
Benzene
Toluene
Xylenes
Required Concentrations (mg/l)
Before Treatment
1 25
30-250 10 30 25 0.01-5 O.QO5
After Treat men t
0.01-5 0.005
0.01-5
0.01-2
0.01-2
0.005
0.002 0.002
0.01-5 0 . 0 5
20- 100 2
50- 100 10
DF
0. 25 1.2 1
1000
1000
1000
1000
1q08
50 10
53
ORGANICS j VESSEL
1
P
Sampling points are also placed at various positions in the system and these valves, which e used to dmw off samples far analysis of the iwcsrganics, of 12 to 15 ml volunie, which provide samples for 2 liter glass flash which collect samples for PCB
ysis of the volatile organics, and large
The inorganic eclmpnen& of the influent are p r e p ixture is made up by simply mixin
bottle,
The air sparging c o l l u m is made up of a glass cylinder fillled with glass enters the column at an upper side inlet and exits the column from a lower side outlet tube. Ais or oxygen is bubbled through a glass frit at the bottom of the a l u m producing very fine bubbles which assist in the evapration of much of the volatile organic content of the waste stram. The expelled gas i s relased from the top of the open column into a fume hood.
The ion exchange columns are prepared by placing the appropriate material in a glass tube. In later stages of operation of the benchscale unit up to two additional columns of A material were placed between IA, and lC, in the system. Solution is usually allow through the column against gravity such that any gas bubbles formed in the columns do not interfere with the flow of solution. A gadliquid separator which allows release of my excess gas pressure to the atmosphere is also placed in the system.
All tubing in the system with the exception of that used to flow solution through columns 2B, is made from Vitori due to its cheinical resistance to the organic c~mponents in solution. At columns 2D no organics are expected to be present in solution and hence tygon tubing is used in this section. A 5 p m filter is placed In the system so that any undissolved solids in the influent do not enter and clog any of the columns.
A pilot plant was designed and operated at flow rates of around 1 gallon per minute for about 6 weeks. The design of the system is shown in Figure 3a and a view of its location is given in Figurc 3b.
'I'he inorganic influent is prepared in a 100 gallon mixing drum by flowing faucet water, conccntratcd solutions of the inorganic components and 0.2 M HN03 into the vessel. The faucet water passes through solenoid valves at a flow rate of betwen 1.2 and 2 gallons per minute. This flow is, however, frequently stop@ by the water cut off solenoids when the volume of influent in the drum riscs above around 80 gallons. The flow restarts when the volume of influent in the mixing drum fd l s below about 138 gallons. This is determined by high and low level sensor electrodes in the drum. The concentrated inorganic components are prepared weekly in 10 litre polypropylene containers as described in Table I T - These mixtures were ~ ~ ~ t d such that no precipitation woiiitl occur and the concentdons were. chosen so that a
56
Fig. 3(a) Design of Pilot Plant
U
I fPm" LIDU'D VAR:AlION
SCNSCR 1!1 'HIGH/IOW ~ CRITICAL
0RGP.YlC S7RIPP;NG COLUMN
SENSOR
VAPOR
CRITICbL
t
MAIN PEESET AUTO ACIC CHEMIC4L 'NJtCTION 'NJECTION
I --I-- d INFLU€Nl MIX DRUMS A T I00 go1 Pil ' 3 B v2
&" 1
7 DRFl 3 M ACIU 1ANK A t 100 'jal J &4 bCI0 TANK A T 100 go1 : and J N
.JA34TON A T 100 gal I I b i l 3 RLCEIVING IANUS v 1 2 v10 b 9 V6 V? va
ORNL DWG 92A-253
18'--3"
FIXED WORK FiXED WORK BENCH SENCH
P U R G I N G
2 TREATED WATER
v) v)
'3 0
2 TREATED
12-ft SQUARE x 1.5-ft DEEP CONTAINMENT W i T t l i l Q i J l D
DETECTiON rY13NITOR (1 5 ,00@-~a l EMPTY VOLGME)
0 R G k N I C
"'\ --Am -J" E LUT IO M
TANKS AT 100 gcri
/-- TABLE FIXED WORK B E N C H
PILOT PLANT ROOM iNTERiOR
ntinuous flow of 1 ml/min of each of these solutions int ~ n ~ ~ ~ ~ t ~ o n in the influent.
using a peristaltic p trol the flow of a
pH of the influ
e solutions are
WNQ, wiluaion from a 55 g rum into the mixh &ns in the range 4.5 to 5
ing a mechanical s t imr and i s pumped out of th at a flow rate of 1.11 gallons per minute.
A mixture of the organic components, prepared as shown in Table 1x1, is also a rate of about 0.95 mXimin into the system using a peristaltic pump and the organics to mix with the rest of the influent by passage throu~h either a 1/2" or 3/ The results of studies of preparation of the influent in laboratory s a l e ex
e influent i s then dbwed to flow through the top of air sparger made from a high glass column with a stainless steel and teflon base. Air is passed into the column from air compressor at a flow rate of typically 9 to 12 cubic feet per minute although replacement the air compressor towards the end of the mn allowed flow rates of only about ti to 7 cubic feet per minute. The air enters the sparger through stainless steel piping with many holes of about 2 min diameter drilled through it in order to increase the surface area of air exposed to the ~ o ~ u t ~ ~ n . .Level sensor electrodes are also placed in the air sparger in order to control volume of water in the vessel to between 3 and 4 gallons. The solution is pumped out of bottom of the air sparger by pump P2 at a flow rate of 0.8 to 1 gallon per ~~n~~ uhich is always lower than the flow rate of water into the sparger. Thus pump PI was designed to shut off when the level of solution in the sparger reaches the high level Sensor and res^ M'
falls bdow the low level sensor.
The influent is passed through a series of columns containing isn exchange rn top to bottom in order to reduce pressure build-up in the system. "lie design of a shown in Figure 4. 'The first 3 columns were made of stainless steel becaust: of the resistance to attack by organics while the remaining columns were manufactured from PVC or transparent plastic. All the c~ lumns have a 6" internal diameter and a height of 6 feet, They are capped with flanges to which 3/4" piping is connected. Piping connecting the first three columns is made from stainless steel while PVC tubing is used in the remainder of the system. All the piping is joined to sampling points using "quick-connectors- which allow convenient by- assing of columns when required. Sampling points labelled V-l through V-11 are placed before
after every column in the system.
Ion exchange rnedia were p l a d into the columns to the required volume and backwashed wit11 faucet water. Samples from the pilot plant were taken on an ~ ~ ~ r o x ~ ~ a ~ g ~ y daily basis by opening the valves at the sampling nts. releasing a smdl volume of liquid into a radioactive waste container in order to clear ti Samples for volatile organic analysis samples were collected in regular vials.
llecting the appropriate volunie i in specially sealed vials, while
59
-..I____c .....I-__
_I._ .̂..._I_
._.._.._..........I. _I_
-.- .............. -.--
a All solutions are made up in 18 liters of solvent
Tc solution is made with a 8.M mCi/cm3 stock sohion
Table 111: Preparation of Organic Mixture for the Pilot Plant System"
... -_ ........
T0luelstZ ....-
' Mixture is piepard weekly im a glass bottle
60
...
...
Fig. 4 Desi CO~UIWIS for Pilot
5”-0”
I
i
3/32” Cryp,)------- I
ALL MAT2 S W L 8E 304 S.S. UNLESS OTH€RWlSE NOTED
61
A system for stripping u mlumras was also built for the pilot plant although not u in this project. It consists sf a series of column for holding media to trap the stri
gallon tanks for storing acid mdicwtive ~ ~ ~ ~ n ~ ~ ~ .
d a similar BVC
The complete pilot plant system i s contained wi n frme covered with laboratory. A liquid
aulh in order to
flcwr of the csntainriienit area, which shuts off the whole system in Tne to*d volume of the cnntainmerit structure i s 15 than one day's flow of water in the event of leakage.
Samples of effluent from the columns, which are eakm at prdetermin mmiially or using an automatic sampler, are analyid using a variety of techniques. The irrductivcly coupled plasma-mass spectrometry (ICP-MS) method is used for determining concentrations of inorganic ampnewts in solution, with de trillion for ,wine elements. The D@ plasma instrument is which cmnot be monitored easily an the ICP-MS. These elem , IC md Ca. The organic cmmyonents ill s d tion are monitor by the gas chroomatogmphy-mass spectrometry (GC-MS) technique:. ]>ah reduction of the collected analytical data is camid out in Lotus 1,2,3 spreadsheets which acccarnpmy this report.
on limits as low as 10 parts
Samples were prepared for ICP-MS analysis by dilution of the sample until the corrcxxtratiabns of all elements to be analyzed were less billion, with the exception of &sodium, magnesium aid calcium. Dilution H N Q with the exception of certain samples from acid stripping experiments where deionized water was u as the dilution rnedialm,
FOP every analysis mn on the samples the response w e of the instruments (res ppb of Li, Mn, I s of interest were then
e made from 1% FINO3 was
ounts versus mass of isotope) was determind using a sa arid IJ. The signals obtained in dnill regular sanipples for
autonnatidly corrected for the mass of the element. A blank added to every analysis run and all the samples in a runy spiked with 25 ppb In. 'The data for all samples is then instrument variation according to the signal for In''5. All ICP-MS data given in this report are
Analysis by ICP-MS of the instrument scans a prdeter the rmge re;lchhg the detector sensitivity required, Tc-99 was
d in a scanning m number of ions of each mas in
e of the low levels of Tc in solution and the high separately by operating the ICP-MS instrument in a
e in which the detector counts only the ions of mass 99 arriving at the detector.
62
_... .
CB’s in the samples were ma4 traction according to EPA rn
between 5 and 508 mls of the analytical samples as an internal s extraction rpf the PCB is is then dried with argonI tQ 20
into 2 mlls of methylene ChIOl-4
and 2 #d Of this SOlUtiOll is inject& into the Same gas Ck The injector ~~m~~~~~~ is for 1 minute. The colsarn
y remains at that tern
For both volatile organics and PCB’s, standard res compound of interest by measuring their responses at
for a
concentrations of e o r n p u n b in samples are then determined from these curves,
‘Test PI
f i m t s to be performed in
(i) Study of the individual stages in my complete g r ~ ~ ~ d ~ ~ ~ ~ scheme. Figure 5 shows how the complete benchscale scheme w the fo1lowing stages:
1. Influent preparation 11. 111.
Air sparger to remove la^^^^ organics Column IA, to polish the water from organics and to remove Tc
IV.
V.
Columns lC, and IC2 to remove U an Crg)i, and ka act as a p for Column lA, Column l A z to act as a backup column to ensure that no radi organic materials s into the f ind stage of the system
to remove the remaining inorganic toxics kern the
12 test runs were mrried out in this project including the integra scale runs in order to chmcterizx the optimum performance sf the d Statistid crptirniation of the system was pdorrnd by ssrnethes OIP columns, and by determining the certainty of the results from th the s p i e s under investigation.
(ii)
(iii)
(iv)
Construction and investigation of an inkgra'ateb system in order to check that all the individual columns will pe connected in seriess. One run of the complete system wa about 100 days. The system ran essentially csntinuall except when routine maintenance, such as replacement of columns, took place, Sampling occuxad usually every weekday although samples were sometimes taken more frequently especially at the beginning of the demonstration and less frequently towards the end of the run. Each ample was 15 rnl samples in the set up were removed in specially seld vials such th organics would escape. 580 ml samples for PCB analysis were taken from the appropriate effluent container in the system, The experimental parameters of the run are given in the Experimental Results section of this report.
Construction and operation of a pilot plant system in order to demons decontamination process the pilot plant was made
also operate in a large flow rate of0.8 to 1 g
unit. Again m e run of per minute until around
allons of wastewater had pas through the system. Sampling usually once every weekday Athou ditional samples were taken an some
weekends. Sample sizes were the Same as in the benchscale unit ex- for those required for PCB analysis when a 1 litre sample bottle was filled for es with low concentrations of PCB's expected.
Study of methods to eliminate the mixed waste produced as spent ion exchange media in order to make their disposal more economically viable, This was performed on columns already used to study an individual stage of the nation process or in the benchscale run of the integrated system. Since the capacities of the columns were found to be so high no appropriate spent columns were available for study at the end of the pilot plant run.
Typically in this project the ICP-MS was set to yield data for inorganic elemz detection limits of around 1 ppb after the sample was diluted. The accuracy of the data is around
64
. ... f 45-20%. In the detection limits of 5-1
e results are accurate to within about e logarithms of large ~ ~ ~ ~ ~ ~ ~ a t i ~ n
errors in analytical data yield con ~ t y assurance samples were nm on ding otp sample q u ~ ~ ~ y in order to
in many of the Figures in in the w h m n characteristics.
ple results are given
Many r n ~ s u r e ~ e ~ ~ ~ of Na, g, Cu are Jn a non-linear range of oul therefore not $e treated as q u ~ ~ i ~ ~ ~ v ~
average relative standard d e ~ ~ a ~ o n of the data for all the elements in these samples is around 19%.
The results for 1 the different sections of the test plan are described below.
A. Individual Stages of the Process
L Preparation of Influent
Several difficulties have to be overcome in mixing’all the components of surrogate groundwater. These include the S ~ Q W dissolution of organics in the water mixture wkich w1 lead to their volatilization before they are completely dissolved in solution and precipitation of several of the inorganic components on mixing together. For example r o d , B a s 0 4 and ZnCrO, are all insoluble in water as are many iron salts at the pH levels desired (4 < pH < 7).
The problem of dissolution of insrganics was overcome by preparing two influent solutions, One influent solution was prepared for testing columns IA, IC and the integrated
.....
system while a semitd influent con for testing column 2113 which was
ticm sf the first influ
11.1
17.1
25.7
23-9
17.6
16.5 .I
Water
h h
10 0.1 50
0.1
10 2
3Q
Q.01
2
5
10
0.1
1
38
Max
1
1
so 1
100 2
30
0.81
2
5
10
1
1
250 30
Tap Water
0.032
28.7
0.005
0.018
5.48
6.57
0
15
10
30
0.032
1
1
10 0.018
0
0 6-57 0
1
15 228 +
iron salts are insoluble at desired levels.
68
Table V (continued)
Species Concentration (rng/l)
Desired Surrogate Actual Water Water
Organics Min Max
Carbon tetrachloride 0.01 5 1.6
Tetrachloroethylene 0.0 1 5 0.2
Vinyl chloride 0.01 2
Methylene chloride 0.01 2 (4 Benzene 0.01 5 1.8
Trichloroethane 0.01 5 0.4
Toluene 20 100 70.6
Xylenes 50 100 41.7
PCB 1 1 0.7
(a) Concentration of methylene chloride cannot be determined since it could not be retained on the GC-MS column.
69
J Y
sf Mixed Waste Groundwater Showing Zndividuai Stages
111
I i R
'6
RATOR
I A ! EFFLUENT C D r l T A i M R
U
V
1C2 EFFLUENT CONTAINER
3"- 1
LEGEkD '- - PbMP
P: - AS NECESSARY
Table VI: Influent for Column 2D
...
P 2
Technet iu rn-99
Influent is made up from eionizd water.
71
The principal ex ri8nent-d pmmeter which was vari the ratio of flaw rate of air in Table VIIa, 189 in Table mlumn was also varied as e results in Table VIM.
to flow rate of influent. Its value was 938 in the ex VI% and 22 in Table VJk . Tie level of influent i
om campasirag the data from Table el of influent in the air sprrrger by ab
species except mrbn tetrachloride by less than 15%, which is close to the ~ G C H
for the organic analytical data. Passage of the influent through the air stripping column at the two highest &ISW rate of
air to flow rate of influent ratios appear to give similar results, DFs of around 1OoQ are achieved for benzene, toluene and the xylenes, while the value is much larger for arbon tetrachloride md about an order of magnitude less for tPichloraethme, Passage of the influent through colrimn 1A is then very effective in polishing the solution yielding levels of the organics in the f ind effluent of k10w 1 ppb for d1 the comporients except kichlorocthane which is just a little above this value. It should be n owever that the level of trichloroethanc is an order
itude higher in the influent in these exprirnents than that required in the ter and sa the DFs for all the volatile organic components are clearly above those
required using an air stripping column - column 1'4 set up under these conditions. At the lowest flow rate of air ta flow rate of influent ratis irsed, the results presented in Table MII (c) d i ~ ~ r a ~ ~ ~ & ~ i t e that dcxontamination factors of only about 10 to 60 are achieved using an air stripping column. The sub,quent C Q ~ U I T I ~ 1A however also reduces the level of all the organics to bclow 1 ppb.
The optimum conditions for operating the air stripping column appear to be close to those given in Table VI1 in which the flow rate of air to flow rate of influent ratio is less than 200 an the height of influent bubbled i s under IS cm. Any increase in these values Qws not l a d to any
onate increase in the DF's md under these conditions around 99.5% of the volatile organics are remove41 from the waste stream thus prolonging the lifetime of column 1A.
111. and V. Column 1A
f column 1A to rcrnovc volatile organics from the influent was described in , while its ability to remove PCR's and toxic and radioactive inorganics ere in experiments run until nearly 1600 column volumes of solution were
passed through it. Table VI11 describes data prdircecE from an experiment in which influent is passed through column 1A with a residence time of 6 minutes. Figwe 6 shows how the logarithm af the dmntarniination factor for the various ions va-ks witla the number of coiurnn volurnes of influent b9at pass through the column. It is clear that the column is excellent for
72
Table VII: Air Stripping Column 1 Column 1A
. . .,
a) Column parameters Air S i t r i w
flow rate of air (F&) = 816 cm3/min flow rate of influent (F*) = 0.87 cm31/min F,i, /F, = 938 height of influent bubbled = 16 cm
Column 1A flow rate of influent = 0.87 cm’/min column volume = 6.0 cm3 residence time = 6.9 min
Carbon tetrachloride 1.5 < 1 >1500 <1 > 1500
Tetrachloroethylene < 1 < 1
Trichloroethane 45 330 136 1 330 4SOOO
Methylene chloride
Benzene 25 20 1250 < a >20 >25000
Toluene 108 120 833 < 1 > I 2 0 > 1 m
Xylenes 15 1s IO00 < 1 > 15 > 15000
Note: Vinyl chloride, which is a gas was found to be too volatile to remain in solution as appears also to be the case for tetrachloroethylene. Only tentative andytical data mould be obtained for methylene chloride suggesting values of its DF in the air stripper of around 3 orders of magnitude.
73
*Table VII: Air Stripping Co'luma / Column 1A (coni)
Colurnn 1A 0.53 cm'/min 6.0 c1n3 11.3 mi^
- flow rate cf influent - wlumn volume - residence time -
- I
Carbon ee:rachloride 5 < 1 < 1
Tetrach~oroethylene < 1 < 1 < 1
Trichioroethane 75 1855 48 6 309 12
Methylene chloride
Benzene 75 65 1150 < 1 >65 >75oeK1
Toluene 215 300 7 LO < 1 >308 >2f5000
Xylenes 70 40 1750 < 1
Note: Vinyl chloride, which is a gas was found to be too volatile to remain in solunioo as appears also to be the case for tetrachloroethylene.
...
c) Column parameters
23.5 cm3/min
22
- - flow rate of air ( F ~ ) flow sate of influent (Fa) F& Es - height of influent bubbled = 17 Crn
1.08 Clm3/KIIiA - - -
Column I A - flow rate of influent I
coium volume - residence time -
6.0 cm3 5.55 min
- -
Carbon tetrachloride 8.5
~ ~ a ~ ~ ~ o r o e ~ y ~ ~ n ~ < 1 < 1 < 1
Trichloroethane 75 3500 21 < 1
ylene chlsride
Benzene 60 5808 12 < 1 ss Toluene 165 3 55 < 1 > 3
Xylenes 25 1500 17 < I
Note: Vinyl chloride, which is a gas was found to be too volatile to remain in solution as appears dss to be the caSe for tarachloroetbyIene.
75
Table VU: Air Stripping Column ! Column 1A ( c ~ n t )
d) Column parameters Air S t r i m
23.5 cm’/min
22 25 cm
- - flow rate of air (F&J flow sate of influent (Fa] F, / F s - height of influent bubbled -
- - - -
Column 1A - flow rate of influent -
column volume - residence tine -
6.0 c1n3 5.55 min
- -
Species Conc in Conc in DE: Conc in DF DF influent effluent air efluent 1A Total (PPmr of air stripper of Col 1A
stripper @Pb) @Pb)
Carbon tetrachloride 1 15 64 < 1 > 15 > 1
Tetrachloroetliylene < 1 < 1 < 1
Trichloroethane 80 3580 23 < 1 > 35 > 8
Methylene chloride
75 6008 12.5 < 1 > 45008
180 3ooo 60 < 1 > 18
30 1580 < 1 > 3 m - 20 ~ .- _..__-_.._I-
Note: Vinyl chloride, which is a gas was found to be too volatile to remain in solution as appears also to be the case for tetrachloroethylene.
76
Table VIII: Column 1A
Column parameters flow rate of influent = 0.5 ~ ~ ~ / ~ i ~ column volume = 3cm3
= 6 min
TC 0.0012
Mi!? 40
@a 50 Na 6.5
K 3.6
Cl- 15
N 0 i 228
SO,- 30
COLUMN IS VIRTUALLY TRANSPA
TO THESE IONS DURING THE RUN
PCB 4 < 1 r500
' Number of column volumes of influent passed through column before DF falls below 100 for Cu, Cd, and U, 5 for Cr, 40 for Tc and 4000 for PCB.
77
4.5
4.0
3.5
3.0
2.5
2.e
1.5
1 .O
0.5
0
-8.5
-4
-Mg
0 -Cd 0 -Ca
A - C r -CU
v -u
0 8.2 0.4 0.6 0.8 1.3 I .2 1.4 -1 .6
NUMBER OF CQLLJMN VOLUMES (THOUSANDS)
...
...
dix A are su~nrn
clearly the most successfuuI material
~ l ~ o s t directly through t candidate for effectively
1. Benchscale Unit
A benchscale version of an integrated groundwater d ~ o ~ ~ i ~ ~ ~ ~ ~ syste and operated based on the data obtained for the individud stages. The h d set up of this un
re 2. It consists of one air spaaging cslum e the first air stripping column re~noves a very large
sufficient to remove the VBC’s to the low 1 hence column lA, is placed after it in the system to
PCB’s and Tc. Thus no Tc car organics are expected to pas and 1C2 can remove the remaining uranium and chromate
79
Table IX: Column 1C
Column eters flow Pat fluent = 0.5 cm3/rnin column volume = 3cm3
= 6min
cu 10 10 1088
Cr
Cd
1 50 2Q
1 1 10 ~
I j IJ 1 0.3 3m
TC 0.0012
Mg 30
50
Na 6.5
3.6
C1- 15
228
30
4
200
> 4
COLUMN IS VIRTUALLY TRANSPA TO l-l-EsE IONS DURING THE R
* Number of column volumes of influent passed through column before DF falls below for Cu and U, 40 for Tc, 20 for Cr and 4
80
4
Fig. 7 DF's of Cofurnn fC ORNL DWG 92-248
I I I I I I i I I I I I I I I I
0 0.2 0.4 0.6 1 .o 1.2 1. 1.6
DF Capacity (Column volurnes)
10 I
10
10
2
30
50
7
2
15 3
8.2 8
< 0,025 900
6.7 1500 > 1200 4.0 2508 $00
30 65
?hnmber of column volumes sf influent ssed through d u m n before DF falls l@i for Ba, Gd, Cu, and Zn, and before DF falls below 10 for Fe,
82
5
4
3
4.0
3.5
3.8
2.5
2.0
1.5
1 .Q
0.5
0
-0.5
ORMk DWG 92A-249
A 87 -Fe
A -K V-Be
I I I I I I I I I
0 200 400 600 800
NUMBER OF COLUMN VOLU
.....
and may also pick up some quantities of other toxics such as Cu and Fe. Thus no ~ Q K E
radioactive material is expected to be present in the solution beyond the 1C columns. As insurance against leakage of either organics or radioactives beyond these columns, colwnn I4 is placed in the system after the 1C columns. The two 2D columns placed at the end of system can then remove all the remaining toxic inorganic cations. These include eopper an iron, when they break through the earlier IA and 1C columns, cadmium, and if they may be present in the groundwater, barium and zinc. While the benchscale system always included an air sparger, A columns, C columns, an A column and D columns in series they were conhUaaly being exchanged as they lost their ability to remove contaminants from the water and furthermore additional columns were sometimes introduced or removed in order to check various characteristics of the system. The typical operating parameters of this system are given in Table XI and provide a 6 minute residence time for passage of influent through each of the ion exchange columns. The sampling schedule in Table XI1 was planned although initially mare samples were taken for analysis in order to ensure conect operation of the system.
Analytical data from the first week of operation of this system are given in Tables XIII (a), (b) and (c). In Table XI11 (a) all the toxics and radioactives were checked before the air sparging column, after it and after column 1A. As expected the air sparging column substantially reduces the concentrations of volatile organics compounds in the water and they are all removed to concentrations below the target values by column 1A. After 1 week of operation column 1A essentially still removes all the technetium and uranium as well as PCB’s, copper and most of the iron. A substantial fraction of chromium and cadmium pass through this column in accordance with previous data. The major non-toxic inorganics such as Na, Mg, Ca are also not collected by the column although fluctuations in their concentrations are observed when one majority ion replaced another during the run. Determination of the relatively high concentrations of these majority ions did not produce accurate data since they fall outside the optimum range of measurement by the ICP-MS. Data for Na, Ca and Mg should therefore only be considered as qualitative.
The behavior of column 1C is shown in Table XI11 @) where the concentrations of the remaining toxics and radioactives in the water before column lCI, between column IC1, and lCz, and after column IC , are monitored. Again in accordance with the earlier data it may be observed that the chromium is picked up very well and the copper has also not yet broken through these columns. The cadmium level however essentially remains constant and is not affected by this column. The fluctuations in the cadmium levels given in Table XIII (b) fall within the error of measurement by the ICP-MS. The cadmium species are, however, removed from the water very efficiently by column 2D as may be Seen in Table XI11 (c).
The benchscale system continued to operate successfully for a period of nearly four months. The capacities of the columns in terms of column volumes of influent passed through the column are given in Table XIV for the A, C, and D ion exchange media. This information represents the whole column run except for the final 3 days when regenerated columns were placed in the system.
13.5 crn - - Height of Liquid in Air stripping Csaumnm
86
Table XII:
......~
2 x 15
2 x 15 I-A, I-Tc
5
5
(1-2) x 15 Daily I-A, I-Tc*
(1-2) x 15 Daily I-A, I-Te*
(1-2) x 1s I-A, I-Tc*
5
5
15
if PCB's in S,, > 1 G- PPb
's in s3 >. 1 PPb
DajlJJ I-A
G-
15 Daily I-A
15 Daily 1-24
G-v = GGMS (voc), G-P = GCMS (PCB), I-A = ICPMS (all eiems), r-%c = ICP
* Only run I-Tc if the Te level in S3A > 10 ppt
...
87
Calcium
........
Soili~irn
CCP, 0.05 - .
0.8 -_II -- .......... .............
5'7500 I
... Table XIII: A n a l y t i d Data from Integrated Benchscale Water Bxontaminati System from First Week of Operation (cant.)
b) Column 1C
Species I Concentrations at Sampling Points (ug/liter)
s4 s5 56
Uranium-238
Chromium
Copper
Iron
Cadmium
0
3%
5468
147
1067
0
7
29
115
1076
0
4
0
105 2095
Table XIII: Analytical Data from Integrated knchgale Water Decontamination System from First Week of Operation (cone.)
-. . ._. .. - __..--._I
----- 111111 ......_... _-.
Concentrations at Sampling Points (ugllikr)
s7 S8
5 13
84 4 4
5 80 1
TABLE X W : Golu'rnm Capacities Determined from the Bench
while column 1s in wsihon
1 A3 1 2564 I 4 2 I5768
... . . 2643 2640 5283
2638 2900 5538
13596 I 4 4 5 I4041
11 2D10* I I 4 4 5 e i I I
91
,411 the A cnlun-rns with the exception of the 1A, column were pla at the front end of the series sf ion exchange columns, Up to three 1A mh in series in th positions. The first sf the thre columns renioves the rqan from the influen while tlPe second d u n i n acts as a backup. When the first cslurnn loses its ability to sufficiently remc?ve the cmntaaminxnts it may still remain tern md be partially effective until its decontamination factors become close to 1. The the water md the first column is efficiently u il it is fully load A media a third column was added in the latt the column before the organics and it was useful to a1.w h o w the -pa organics since i t may make the stripping of the columns more efficient.
d solumn in xieries con~nues to YpYQCeSS
Column Ih, was initially placed at the front of the ion exe ge system immediately after the air sparger and it remained in that position until Tc broke ugh the column after 6615 column volumes of influent passage when i t was removed from the system. Columns lA, while placed initially in a backup position was also allowed to remain in the system until Tc broke through and was in the front psition for passage of 42@4 c o h m n volumes of influent. Thus an average lifetime for an A column can be estimated to be 541 V's based on Tc. Column l& was placed at the front of the system after being in a backup tion md was allowed to remain in the front position even after i t was fully loaded with Tc. It remained in the front position until the end of the run yet it continued bo efficiently polish the organics from the system. Thus the capacity of a 1A column for renwval of organics remaining in the influent after air sparging is grates than 98800 column volumes.
'I'hc IC columns and the l A , column placed after the 1C columns were found to be fully effective for the whole duration of the run and it is therefore clear that these columns can remain on line for at least 20,008 column volumes.
The 2D colun-nns were located at the end of the ion exchange system and a pair of these columns was always in place with one acting as a backup for the other. When the DF of a toxic metal contaminant such as copper or cadmium fell close to 1 then the front column was removed, the backup column took its place and a fresh column replaced the previous backup column. Ten 2D columns were re4uired for the whole mn with columns 2D9 and 2D,, remaining at the end of the run. 7he average number of column voluiries of influent that pas between exchanges of columns is determined from the data in Table XIV to be 2386 734. The large standard deviation in this number is partly because some stripping solutions were added to the system during the run and i s partly due to the fact that analysis was performed only five days a week and a lag time sometimes occurred between sampling and the decision to replace the column. This in no way affected the quality of the effluent of the system since a backup column was always in place, but a large time lag would extend the time a column in the front position would remain in the system and reduce the time a backup column would finally remain in the front position.
The data obtained froin the integrated benchscale decontamination system clearly shows that the mixed waste contaminated influent is cleaned using this ion exchange technology. High
92
... .. ..
...
capacities of the columns have k n achieved and scale up of pilot plant study discussed below.
The pilot plant described earlier was operated for a period of &QU
continually except for routine maintenance and when problems were gallons of surrogate water were passed through the system and it not o at processing the wastestream but it also yielded experime results obtained from the benchde studies, A similar seri columns to that in the benchscale study was set up although only ei each type was in operation at any time.
Initially a flow rate of one half of a gallon per minute influent into the system and this was gradually increased until the flowrate was between 0.9 and 1.0 gallons per minute. The flow rate of faucet water in mixing drum and the flowrate out of the drum into the air sparger were obviw so that the influent mixing drum and air sparger did not drain an Careful manipulation of these flowrates was necessary since Q shutoff because the level in the air sparger fell below a critically low le the influent mixing drum overflowed since the flowrate of faucet water high and the solenoid valve controlling the flow failed. The opthum this system was found to be 2.0-2.5 gpm for the faucet water flowing i gpm for the influent flowing into the sparger and 0.85-0.95 gpm for th sparger into the ion exchange system. The pressures at the individual columns daily. At the highest column load of six 6 cu ft columns and at a flo the maximum initial water pressure was 120 psi with a pressure drop of the initial 1A columns, 22 psi over each of the subsequent 1G and i A ml psi over each of the 2D columns. A pressure Sensor after the 1C col down if the pressure at that position surpassed 46 psi was never acti run a
The flow of air into the sparger was initially held at the desirable levd sf I. 1-1 1.5 cfm but this had to be reduced to 9.5 cfrn because of the heavy load on the compresfor. For the last twelve days of the run a new compressor was used, however, it could only deliver air Rowrates of 6.5 cfm. Although this led to less efficient removal of volatile organics f m the system the first i A column was found to be very effective at polishing the remaining corn
The flow rates of toxic inorganics, radioactives and organics were maintained at a steady 1 ml per minute. This was calculated to yield the required concentration of influent when the total flow rate of influent through the system is 1 gpm. Fluctuations in influent concentrations occurred because of changes in influent flow rate but these were monitored daily by sampling at point VI. The pH of the solution in the influent drum was set to 4.7 but varied between 4.5
93
and 5.0 because of the lack of sensitivity of the pW monitor. This pH is a little lower
sparger or elsewhere in the system, Indeed no precipitation was faun the whole run
n c h d e set-ang but was chosen ts ensure that no precipitation
pair-its between VI md VI 1 QX% m approximately daily basis. A typical from the pilot plant taken about 2 w ICs after the mn began is presen @onmesatrations of Samples at mmplin pints V1, V2, VB, V7, V8, V1 after the mlurnns shown in the table are tabulated in units of ygll for dl s md Mg which are in mg/l. T%e reinaining sampling pints were not in during the run,
.The measured concentration of the initial influent entering the s It was attempted to produce m initial influent with the sane cawatrations md radioactives as those given in Table I1 with the ~ x ~ ~ ~ ~ n of CQ ~ ~ c e ~ ~ ~ ~ a t j ~ ~ af 5 rng/l. It was planned to place 5 ppm each of dl the volatile org xylene and toluene which were set at 100 ppm, PCB's were not added to of their lack of availability but small amounts were added on one day of the run so that the effectiveness of the system to remove the could a h be dcterrnind.
The actual eoncentration of influent did vary somewhat from day to day but typicdlly the toxic inorganics and radioactives were within 3Q% of their desired values while the organics were often considerably lower. For exanpie in the data given in able XV the can
riics msasu rd in the influent of V1 are less than half the eicipatd values nd quite well with that given in Table XIII for analytical data fmm the
e system. The toxic metals including bariuni, cadmium and ion exchange series until they reach the 2D columns where they ahe ve
T" also loads the 1A and 1@ mlumns and in this example I1 taking place resulting in a DF of about 9 for Cu on this
to polish the leve:s remai very effectively picked up by the IA, column a d most of the uranium is dm l A , mlumn but the 1C column is n a x s also requires a lC column for polishing the levels remaining after passage through the IA, mlumn.
"he levels of organics in the influent were lower than antici mixer in the pilot plant i s less effective than the influent mixing system for mixing organics into solution. However in spite of this fact the data for in Table XV for the pilot plant again correspond well with the data obtained from the system given in Table XBII. The air spuger reduces the concentrations of V W ' s in the influent by factass of between 16 and 55 for the chloro substituted eompunds arid between 9
94
9
Q 45 _lll_
24 0
92
e, a 5858 I 6751 734
. . ...
43
3
Trichlomethanr I 0 0 I O
0.082
2575 0.058
95
the aromatics. These DF’s we a little inferior ere probably due to limited air flowrate and g
column is however very effective at polishing yielding ~ ~ n ~ ~ l y undetectable levels of the VOC’s.
A check of the ability of the system to remove PC the run of the pilot plant. Data from this experiment ~R~~~~~~~~~ of PCB’s were entered into the system and a pun
. l (nmnA,40Qm gmic mixture and
PCB’s of I ppm in the influen In run B, the VOC mixture in order yield a co h run the solution was allow to flow for 1 hour before samples were taken md then ana lyd . In the V1 samples only very low levels of PCB’s were found, up to 208 times less than expected. This is attAbuted to dissolution of the CB mixture into the aqueous solution by the static mixer. passage through the air sparger considerably more of the material is dissolved, the high air flow enhances the mixing of the solution. V3 samples taken after influent through lack of ability o the lack of dis.wlution of the PCB’s in the influent. PCbs’s may be present in the influent which only slowly the system. This is demonstrated by the additional d concentration show an additional DF of between column. No are detected in the eftluent of the V11 md thus even in these experiments the syst PCB’s 20 below the required levels since the fin 1.4 columns following the first two 1A columns.
f 1 rnl/min and of influent of 1 gpm.
rst 1A column show removal of over A materid to pick up Pes’s in these
e capable of removing all the s are removed by the 1C md
ibies of the columns in the pilot plant were d
f column volumes of influent passed are given in Table XVn. Si
from the flow of influent of introduction, exchange and removal of columns from the sy
rated for around 40,OOO gallons of contamina water, only values for the lumns could be calculated with any degree of nty. The 1A, column in f the ion exchange system was effective for the entire duration of the run
although on the very last day of operation, sample #43 shows that technetium was finally breaking through the column since a DF of only 2.3 was achieved for Tc. The lA, column only servd as a backup for l A l and the 1A5 column located after &e. 1C column also the end af operation. The 1% column was only of half size and was initially pla 1C column. It was replaced with lA, when additional 1A matefial became available.
lCz was initially plac in the system but when this w z found to be ine with a freshly prepared IC1 column. This column remained in place until the end of
the run and as expected from the benchscale results has a high capacity. The 213 coiumns were
96
Table XVI: Check of Decontamination of PCB’s from Surrogate Groundwater in Pilot
.-.....
97
Golumnm
* Column was still. in o ra&ion at end of mn,
98
....
....
ies added to tap wa
99
le XVIII: Com f Column Capacities in the knchscale and Pilot Plant Systems
100
Table XIX: Waste Loadings in g on Columns at End of Pilot Plant Run'
Column # 1Al* lA2* 1A4 1A5* Size (gals) 6 6 3 6
Barium -0.86 Cadmium -1.49 Chromium 117.87 Copper 12.56 Iron 5.23 Zinc -4.11
Technetium - 99 O.lOoO8 Uranium - 238 99.06
Carbon tetrachloride 16.60 Tetrachloroethylene 8.73 Benzene 33.50 Toluene 178.33 Xylenes 330.34
0.58 -2.10 4.47 75.71 -4.62 1.73
0.0094 1 46.95
-0.12 -0.10 -0.05 -0.61 -1.74
0.01 -0.52 -0.17 12.61 -9.64 -0.29
-0.oooO6 -0.02
0.00 0.00 0.00 0.00 0.00
-0.20 -4.87 -0.02 46.55
2.25 0.05
-0.00050 0.10
0.00 0.00 0.00 0.00 0.00
...
101
Column R 1c1* 1@2 S i x (gals) 6 4
Bariunr -0.16 0.14 Cadmium 1.83 2.28 Chromium 0.24 0.75 Capper 252.00 1’73.24 Iron 1.55 4.81 Zinc 2.24 0.27
Technetium - 99 0.0179 -0.00039 Uranium - 238 0.08 1.19
. _. ........................... - lllDI ..
Colurnn # 2D 1 2 8 2 2D3 2D4 * 2D5* Slze(gak) 6 6 6 6 6
Raxi (1 m 2.31 1.19 1.41 1.51 0.88 Cadmium 11.22 42” 19 15.47 40.43 0.01 Chromium 0.11 0.16 0.05 8.12 0.01 Copper 2.06 34.74 5.52 20.72 -0.00 Iron 4.21 3.99 2”89 2.46 0.52 Zinc 0.17 -1.59 - 1.42 0.56 0.03
* Colurnns in operation at end of run
(a) Small negative values for waste loadings are insignificant and should OK treated as zero. They represent cumulative errors from the ICP-MS dctermina- tions.
...
waste e~imin~tion were
column labeled "Ha which selectively removes on e effluent containing only toxics can be returned
....
Two forms of spent columns 1A are lpraeluced from the water ~ ~ ~ ~ ~ ~ ~ ~ ~ a t i o ~ $ y ~ t ~ ~ . Those columns taken from the first position in the system are lo u ~ ~ ~ m , chromate and some other toxic in anic cations such columns taken from the second pasition in system we Psaded wi 1A column except for organics.
Passage of 0.1 N WNB, through spent 1A colurnns led to incsasin radioactive components being removed until a maximum level is m c number of column volumes of 0. I N HNQ required to be passed Ihr these maximum levels, together with the maximum concentrations a stripping of the two columns lA, and 1% from the benchscale system. Column lA, was
lla, with
organics. Large quantities of copper are Seen to be removed from bath columns using O,1N "I3 although very little separation is achieved between the maximum levels of copper and uranium. With the exception of cadmium larger quantities of material are stripped from the column as compared to the 1A3 column. Only very small quantities of Tc are removd column with this stripping solution, The maximum levels of copxx are removed after
initi toxic
removed from the benchscale system after it had become loaded with TG while remained in the system even after it became loaded with Tc and continued to be
of 5 CV's of 8.1N KNO, and it is therefore proposed tha this volume of 0.1 N HNO, in order ti, remove large qu
ms for Stsippin c Columns
i j
.. .
Table XX: Stripping of 1A Columns from the Benchscale System with 0.1 N HNO,
Maximum Levels of Elements in Stripping Solution
I 1%
Concentration
...
105
Sirice the ultimate goal is to separate the mixed waste, stripping experiments were d w performed in which ~ I I G acid effluent of the I A colunin containing both toxic and riidimctive eom~~~r~emts is passed though ap1. additional coluii~p~, named lH, which was checked for its ability to remove the radioactives from the stripping solution yet let the toxics E ~ & R . Column 1H is a charm4 b a . d Dura-C column which undergoes special treatment in QF&I- to improve its selectivity properties. ‘These exprirnents were camid out in a set up shown in Figure I&. 1 I Aroiind 36 miis of 1N “3, is placed in the influent bath. Initidly vdvc V1 is qeii <and valve V 2 is c l o d . ?%e influent is then pumped through a spent column at a rate of 2 column volumes per hour and then through either 1 or 2 1H columns before entering the recycling containcr. Sampling p in ts are Isca t4 after each column where 0.1 rnls of sample miry be renaovd with a syringe. Samples are diluted by a factor of 100 with deionized water and given for mdysis. After all the influent has been pumped from the influent bath, valve V1 is d o and V2 is opened, Pumping is continued and th,a stripping solution is recycled via the recycling container,
Experhenis run or1 this system using 1A C ~ U ~ R S yield emrs analytical data. with 188 times larger than other data given in this report bwuse of the dilutio~ of the samples. In the run on the 1A4 column €or which concentration data is also given in Table XXI 6 colurnrm volumes of 1N HNO!, are passed through the 3% calumn followed by a single 1H column in 8 cycles giving a total passage of stripping solution of 48 CV’s. l%e data clearly shows that except for about the first 12 colurnrn volumes the concentrations of Cr, Cu and Cd are the .same after the 1.4 cdarmn as after the 111 column indicating that the 1FI column is essentially transparent to these species. In the initial two cycles of stripping, Cd and Cr are seen to be adsorbed on to the 111 column but then almost immediately r e l ead bzck into solution. On the other hand the radioactive species, IJ and Tc, are strip from the 1A column mmd then eff~,ti.tive?y aptiired by the 1M coluiinii. These dah may $e seen in a graphieJ form in Figure 12 where the ratio of the concentrations of elements after the IW column to the cancmtm?ions of dements after the I A column is plotted against the number of column volumes of acid passed in the recirculation systcai. ‘4 value of zero indicates that the i H column icmoves all the cornwrlcwt reIashcd from the B & colurnii while a value of one indicates that the 1 I% column i s trarqsarenit to that spxies. Values grater than one icdicate rclase of excess of ihat element
106
# of CV’S Passed Levels of Contaminants in Effluent from Column 1 4 in mg/L
Cr CU Cd W aa: -
Influent = 0.1 N E”, - no recycling
6 CV’slCycle
#afCV’sPassed I Levels of Contaminants in Effluent from CO~UIIM IN, in mg/L
Influent = IN HNO, - recycling - column IH, in cycle
107
from the 11-1 cdsnmn. Agah it is clear that the toxic elements dl essentially have a an H/A value of 1 while the m;nllioar~ive W a d Tc have MlA values of 0. Note again scatter of data i s expected here sinw dII mipies were diluted by a factor of 100 fo The 1H column i s therefore extremely successful at removing solely the radioactive from the stripping solution.
Stripping af the 1A column with 1N HNO, in this run initially 1
passage of about 4 CV’s of 1N H N 0 3 and then after 48 CV’s the motin column ste;rd;ly decreases to about 1 mgfl while the Tc level remains at
was decided to IseintrcPr’iuce it in the benchscale s regeneration, The final three days OF the k n c h s d e operation were thus run with eke paatidly stripped IA, column in he. front position. The column however was found not 60 be effeetiive in removing IJ, Tc or any of the tnxics from the wastestream. Decontamination facton of less
and less than 5 for uranium were achievd on this column. This suggests that only columns may be regenerated since any waste comp-ments remaining on a parpially n may be concentrated in a band at the end of the column and passage of influent
6.01 mg/l TC md 100 mg/% U FT ~ ~ p l e . Vle level of Tc iach
mgil. Although the column was not compIeleBy strip
through the column niay release waste into the effluent from this bmd.
In order to determine volume of stripping solution required to fully unload to completion in the set up described in Figure 11. column, column 1‘4, was strip
column was used for paysage of the first 60 CV’s md two 1H columns were used thereafter. The results arc shown graphically in Figure 12 in which the concentration of s p i e s in the solution after passage through the first 1H column divided by its concentration after passage through the IA, column is plotted against the total number of column volumes of 1M HNCr, passed through the Id4 columan.
The concentration of uranium in the effluent of 1A falls to below 0.1 mg/l after of 47 CV’s of 1M KNO, in this run and the concmtraticsn of T 1s below its detecti after passage of 140 CV’s. Since 14 CV’s of 1N WNO, was pa down the la, column also in a ~ E V ~ O I J S run it appears that 154 CV’s of 1N HNO, are required to strig the 1A, column of the radioactive components. An EPA Toxicity Characteristic Leaching Procedure (TCLP) test should be performed on the stripped BA wlamn in order to ensure that i t may be classified as nori-radioactive waste, This, however, requires at least 100 g of material and can therefore only be performed on spent 1A columns from &e Pilot Plant. Since the specific activity of Tc-99 is 1.7 x 1Q2 Cilg cornpared to 3.6 x 10.’ Ci/g for depleted U the quantity of Tc-99 remaining on the stripped column can only be 2 x times that of U.
The capacity of the 1H column can also be determined from :he dab., For passage of less than 6Q W ’ s of 1N WNO, through the 1A colutim, comespnding to 46 CV’s of 1N H N Q through the IH colunm, the 11% ~ l t i n i n remains essentially transparent to the toxic cxxnpnewts yet very effcctive a.t rernsvhng ‘Tc and U. At flow volumes above this level lmwever the Tc begins to break through the 1M column as may bc scm in Figure 13. Indeed, for flow sf many
108
Figure 11. Set up for Separating Mixed Waste Produ m
+--------.7 ! i ! I_-------
2 -
1.8
0 0 =
1 . 4
:.2
2.8 7
&
'-
2 .6
2.2
r-id w w WCU 0 0
1 0.6
0.4 1
0.2 t L
0
\
10 20 30 40
D C r
N \ jmber of Cnlclmn Volumes
+ cu 0 Cd A d x Tc
50
8
7 1
6
c d
4
.J
2
P J
Run 199HA - 1 Molar H N B 3 -
\
more cycles of 1N HNO, the 1H column releases more Tc th ever, the: second 1H column in stxi
94 CV’s of IN ” s to break through.
b d1 the Tc stripped from a remove U is howe~er much larger. From this
city of the 1H column for uranium exactly
was dm o p m t d in a run in which 48 CV’S of
of 188 CM’s does some umiuaxa begin to p a
Similar stripping experiments have been per experiments and the bench le unit and the data are mluann from the b e n ~ h ~ ~ d l : system was removed md then with 64 CV’s of 1N HN03 in the recycle sy mlunm. Sample concentmtion data are presented in Table XXH for the ma,
in these stripping exper ents. No Tc was present on the column or stripping data since in the knchs system 1A columns were located prior to order to remove a11 th from the d u m n by
in Table XXIII. Again separation in the number of CV’s sf acid passed betw w the maximum levels of toxics and uranium i s very small. Wow of O.1N “0, removes large quantities of Cu which may be retu
on IC columns from both test irnilar to those for 1A. The le, with 12 CV’s of 0.1M MNQ and n in Figure 111 with just oflie 1W
99. The maxirnum levels of s p i e s sui 8. IN “ 0 3 and the n of evs of stripping solution required to these levels i s given.
removing such high ConEnmtions of uranium.
The data obtained on recycling 6 CV’s of 1N IINO, over 18 times through the setup shown in Figure 1 I are ~~ presented in Table XXII and in Figure 14 which again shows the ability of the ZH e@rlumn to selectively remove U from the recycling solution. All the uranium is renrasvd by the 1H column while except for
rat to all the other toxics. me concent per mlurnn volume of influent i s also found to s ssage of 64 column
of 11% wlurnns for tiraniurn was determined earlier to be 188 CV’s, th is result suggests that one 1M mlumn will he sufficient to remove the radioactive material from a maximum of three 1C columns. It should be noted, however, that the IC, column $1
It3 O f Stripping SQlutiQfl its COnWntKition iS
not fully spent in the benchscale system and T limit. Due to the extremely long lifetime of
this number should be IC columns in the benc
viability of regeneration of the 1@ material was not investigated.
Spent columns 2D are loaded with only toxic and non-toxic inorganics an of these ~ l u m n s will probably neat be viable. Th mixed md xmn ly the column has very high SCdlUtiOll.
because firstly the waste pr ities for capturing the c
112
Table XXII: Typical Levels of Contaminants in Effluent of Acid Stripping Experiments on Column IC
Run #203
...
# of CV’S passed 1 Levels of Contaminants in Effluent from column IC, in m n / ~ 1 I c r I c u ICd
Influent = 0.1 N HN03 - no recycling I
Influent = 1N HNO, - recycling - column lH, in cycle cv’s/@ycle I
1 257 761 11.2 13.0
12 339 197 0.5 10.9
24 405 373 1.7 1.6
1 1 __ ~~ ~-
36 189 268 I 1.3 0.6
48 153 307 1.2 0.2
60 la, 410 1.2 0.1 j
# of CV’S Passed 1 Levels of Contaminants in Effluent from Column 1H3 in mg/L
i Cr cu Cd U i
Influent = 1N “0, - recycling - column IC, in cycle
6 CV’dCycle ’
. . ... .
113
Table XXIII: Stripping of 1C2 CO~IJKHI from Fknch e system with 8.1N "0,
Cd 4 5.1
125.8 - U
114
5
4
3
2
1
0
Fig. 14 Acid Stripping of Column IS! ORNL DWG 92A-247
RUN 203HC-1 M HN03 I
I
I I
Cr 0 cu 0 Cd A U
i I I
20 40 60
NUMBER OF COLUMN VOLUMES
During thz ini t id stages of the oFi;a;ion of the pilot plant the systcm shut i tsel f off tiriles sincc tiic influent in thz air sparger fell below a critical law level. 'This was
adtribt&d to high voltage thctuations that occurred in the pawer supply Que to the extremely hat summer which mused either P1 to pump bo slowly or y2 to pump too fast for a of time, This probkm was overco y increasing the maximum flswmte of influent that may be pumpad by PI. Anotber naajajor em did c 9 c x x T on one of the%e W ~ S i c a t s S when the system
v a s present in the line. This sauwd. the tap water supply to continue flowing system even though the remainder of the system had ceased to operate and stsliciure filled with water until it was discsverd the following morning. Analysis sf the spilt water showed no harmf~il relase of mntamination and the excess water was drained away. An additisnd ~ 0 1 e n ~ i d valve was therefore added on line, as shown in Figure 3a whose function is
ea the system shuts off but ren9ains open o t l w w k The otfier solemid valve 11g m d closing as i t regulates the flow of faucet water into the influent mixing
shutoff and the H,O cutoff mllenasid valve failed to close Ab the time only Q
116
.. .
....
......
's were found to Q n tamhate water
r two four hour
e pilot plant operation ap benchsde run, ~~~~~~~~ ~~~~~~n 0
uired in order fa confirm the bench more exact vdwes for the capacities of the cal n r ~ ~ e n ~ ~ a ~ ~ ~ ~ ~ ~ ~ ~ e ~ ~ s be perf~rned
experiments could be canid out involving study of and investigation of rnehds increase the efficient l ietime of the 1A, column. ch data could lead to even further viability of the decontamination process.
Table XXIV: Estimates of the Number of Gallons of Contaminated FVmssed per Cubic Foot of Ion Exchange Medium
117
As a result of the success of the technology i n v d g a rt in treating the pen water and in ~ e n ~ ~ t ~ ~ ~ no mixed waste, a full-scale plant wuld now k
th contaminated ~ r o ~ n ~ w ~ t e r . The design of such a plant is or flow rates of less than around 10 loris per minute it i s envisaged that
ler which would a l rovide the advantage of mobility be moved from one contaminated site to mother. For plants requiring
ter than 88 gallons per minute, a ial facility may have
In a full scale decontamination system, mixed-waste contaminated groundwater is pum into the system using pump P1 through a filter. Nitrk acid is add4 to the influent in ord bring the pH of the solution to around 4.7. Passage of the influent through a mixer ensures complete mixing of the acid and groundwater. The solutim then passes through an air sparger to remove the volatile organic cornpounds and is then pumped through a series of ion ers much like t h ~ ~ e used in the benchscale and pilot plant ~ ~ ~ ~ i ~ ~ ~ ~ o n units. It is to place 3 1A columns followed by a 1C column, another 1A column and then 3 2D columns all in series. The use of an additional 1A column and an additional 2D column adds some redundancy to the system however these extra backup columns permit less stringent monitoring of the effectiveness of the columns and hence analysis of samples will be necessary only twice a week thus reducing the operating costs quite considerably. Replacement of one of the first 3
perfomed by removal of the spent column m may continue operating
with only 2 of each of these columns while the rep re takes place. However replacement of either the IC, or 1A4 column req f a fresh column before removal of the spent column and hence an extra position between these two columns is available.
s or one of the last 3 2D columns may the addition of a fresh column. The d
water will finally pass through a revase osmosis system in order to tration. The raffinate stream from the reverse osmosis unit consisti of nitrates and calcium and magnesium salts may be used fication process or the ceramic manukturing process r q u
disposal of the spent ion exchangers. Alternatively no reverse osmosis inay be required if the effluent is discharged into a river such that the resulting level of nitrates remains below the drinking water level of 10 ppm.
118
Figure 15 FULL SCALE PLANT FOR TREATMENT OF MIXED WASTE CONTAMINATED GROUNDWATER AND ELIMINATION OF SECONDARY MIXED WASTE
a) Proposed System for Processing Mixed Waste Contaminated Groundwater
F
P1
1A3 lC1 1 4 1
D O 00
4
N 0
1 1 i
1 i l I
1-42!. I T 0 A
-u
...
.-....-
a p ~ r ~ ~ ~ a t ~ volume of‘ the main ~ r ~ ~ ~ ~ w ~ t e r dwon
column, which contains prdominan from the stripping cycle, this IN HN03 is slowly pas will remove the copper and any other loaded toxic cations
be recirculated a column and the captured on a ZH column. The solution in
TG with ScJrne
UandTc. I t m be disposed ~ i f as radioactive was 38 times through a spent I A column inns such that the U and Tc will
3
Figure 16 (b): Proposed Scheme for Use of Nitric Acid for Stripping 1 C Columns
4.67 CVS
0.33 CVS
Discharge Water
I
!
1 \
4.84 w s i
pH 4.5 Ill
to around 5 so that it may then be placed back in the mah system, This tarilk T3 will mntain 6 CV’s of 0.1 N H N Q which magi be passed through a. 1A column immediately after its. removal from the mair-i system in order to remove excess quantities of the toxic inorg the calumn. ‘I’Re effluent of this step passed into tank ‘T4 where it is stared to be slowly entered into thc maire system at p i n t B. The remaining 4.4 CV’s sf 1N acid wlaatisn in tank 72 after the stripping cycles i s uxd for acidifying the i n f l u a t in the main decantamination system by its addition to p i n t A at a rate which maintains a pH of the influent of around 4.5-5.
The order in which a spent 1.4 column undergoes the stripping prw&ure is s h o ~ ~ n by the 6 mlunan volumes of 0.1 N H N Q ai stage I, hen undergms 30 cycles of stripping by 5 CV’s of 1 N HNO, at stage 11, and i s finally waslad with 5.4 CY’S of discharge water at stage 111. Storage of the spent colunins is n m - s q between strippings since the eolurnns will be required in at least two stripping processes for optimum use of the nitric acid in this process k h r e a 1A column will be regenerated.
1C columns cannot be regenerated since their lifetimes in the demntanination process are so long although they can be used for storing some of the toxic anions. ‘rhus 4.4 cdrarnn volumes of the solution in tank 72 after the stripping cycles is neutralized with NaOIQ to a pH of around 4.5 and then passed back through the stripped 1@ colurnn. At this plp the toxic anions, mainly chromate, aic returned to the 1C column and the relatively clean ccfflucnt is added back to the system at point B after storage in Tank T4,
While these stripping processes have not been demonstrated in the pillst plant, since the lifetimes of the columns were tw, long to produce enough spent media in the t h e frame allowed for testing the principal stages have been shown to $e feasible from experiments of the benchsale level. ‘I?hcy are also expected to behave similarly on upsc;nling, Fudermore, the systems showrr in I‘igure 15 enclosed within dotted lines, namely a charcoal filter to adsorb the
TS of VBC’s from the air spager and a reverse osmosis system after the 2D columns to reduce the non-toxic salt content in the effluent, have not been investigated in either our benchscale or pilot plant expcrimenks since they represent existing teehnologies which are readily available corn rnerciall y .
The systems and processes described in Figures 15 and 16 have also beern designed in order to minimize the number of analyses required to be perform to ensure success in processing thc contaminated groundwater. Analysis of samples from the decontamination system will be required twice a week since a back-up ability of at least 4 days is built into the system in the event of column breakdown. Analysis of samples from the acid stripping processes will be required oncc per process i n order ta confirm the sancrxssfiil separation of mixd waste on the columns and the ability of the 169 colunms to perform their f~inctioaa.
A cost estimate ha% been made for the above described system operating at a flow rate of 10,MM gallons per day. ’Three components enter the cost:
124
.... 1. Capital Equipment
A trailer with 12 columns, each 12' diameter, all necessary piping, pumps, level controllers, miscellaneous tanks etc. This is estimated trp cost the custcmer 1.25 centdgallon if the capital cost of equipment is expended over five years,
2. Operator and Project Management Costs One operator working full time with an estimated overtime of 52 hourdyear and me- project manager employed for 20% of hisher time is expected to cost the 4: centdgallsn.
3. Ion Exchange Media and Miscellaneous Chernids Since the capacities of the ion exchange media have not yet been finally ~ ~ ~ e r ~ i n ~ for the pilot plant only a rough estimate for media and chemical costs of 5.38 cents/gallan has been made.
These prices are based on the DCAA audited rates of Duratek with an 8% profit (fee) and assume a 5 year contract after which the equipment becomes the property of Martin M a r i m Energy Systems. These prices also, of course, depend on many other parameters including the waste stream contaminants, the total number of gallons processed and any special construction or operating codes or regulation that may be required to be invoked. Several items are not included it? the cost estimate, these being analysis of samples, disposal of columns and peripheral decontamination systems such as a charcoal filter for the air sparger and the reverse osmosis unit. The total cost of 9.63 centdgallon may be considerably reduced if the contaminad groundwater contains only one type of waste or if the contaminant concentrations we lower those in the surrogate.
....
125
1. "Methods for the Deterrninatbn of Organic Corn aids in Drinking Water", 14-881039, D a m b e r 1988, pa
2. "MeQ~ocIs for the Determination of Organic Cam unds in Drinkjng Water", Number E 9 December 1988, p. 325.
126
127
.- .. .
. .-.
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12" 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29- 30. 31. 32.
R. E. Adam J. M. Begovich W. D. Boshick 9. T. Bradbon-y
T. R. Butz V. B. Campbell R. M. Canon H. M. Clancy 0. K. Clotfelter J. L. Collins K w. Cook A. G. Croff N. S. Dailey T. L. Donaldson H. M. Braunstein R. S. Eby B. Z. Egan R. L. Fellows R. K. Genung L. V. Gibson, Jr. K. W. Glass D. '13. Gunter J. L. Haymore F. K. Heacker J. E. Heiskell, Jr. I. W. Jeter G. E. Kamp T. E. Kent R. K. Kibbe k J. Kuhaida, Jr. D. B. Lloyd
33 - 42. 43.
44 - 53. 54.
55 - 64. 65. 66. 67. 68- 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79.
80 - 89. 90. 91.
92 - 101. 102. 103. 104. 10s.
106 - 115. 116. 117. 118.
119 - 128.
T. F. XDonrlcnick K. Machanoff A. It". ~ a i i ~ a ~ ~ a s N. G. M ~ K ~ ~ c. v. IliS
R. P. J. Q. Miller T. M. Monk M. 1. Morris 9. M. Napier Mom Norris K E. Plummcr W. Poling J. K. Prazniak J. G. Pruett S. M. Robinson D.P .Schae€ferkoc t t er C. E. Scheibly M. J. Shclton R. 2,. Siegrist S. It'. N. Singh R. E. Swatzell D. W. Swindle J. S. Watson E. I,. Youngblood H. K. Yook Central Research Library Document Refer. Section Laboratory Kccords Laboratory Records - RC ORNL Patent Scct - RC EMD Documcnt Mgmh
ER Document Mgrnt Center
Center
EXTERNAI[, DISTIUBUTION
130. W. D. Adams, DOE, FOB, Rm. 2116, Oak Ridge, TN 37830 131. T. D. Anderson, U. S . DOE Headquarters, Trevion Bldg. 11, Suite 400,
12800 Middlebrook Road, Germantown, MD 20874
128
132. D. L. Berry, Saimdia National Laboratories, Dept. 6620, Albuquerque, NM 87185-5800 133. J. D. Merger, Westinghouse Hanford Company, Manages, Office of Technology
134. W. R. Bibb, DOE, FOB, Rm. 309, Oak Ridge, TN 37830 135. K. E. Blake, Portsmouth Plant, P. 0. Box 628, Piketon, OH 45661 136. J. J. Blakeslec, EG&G Rocky Hats, Program, Manager, Nuclear Waste
137. R. A. Carrington, MSB Inc., P. 0. Box 3767, Butte, M?' 59702 138. P. Colombo, Brookhaven National Laboratory, Upton, NY 11973 139. D. G. Cope, PGDP, P.0, Box 1410, 043-1 , MS-l05A, Paducah, KY 42001 140. J. P. Corornes, Ph.D., Deputy Director, h e s Laboratory, 108 Office & T,ab Bldg.,
141. M. Dugger. Kaiser Engineers Wanford Co., Environmental Support Org.,
142. P. H. Edmonds, Radian, 120 S. Jefferson Circle, Oak Ridge, 'Iw 37830 143. D. Emelia, Strategic Planning Dept., Chem-Nuclear Geotech, P. 0. Box 14000, Grand
144. 12. B. Evans, Ph.D., Reynolds Electrical & Engineering Co., Inc., P. 0. Box 9$521,
145. Clyde Frank, U. S. Department of Energy, 12800 Middlebrook Rd.,
. Fj. GEllies, ETEC, D/030, T038, P. 0. Box 1449, Canoga Park, CA 91304 147. C. S. Gist, DOE, FOB, Km. 2116, Oak Midge, TN 37830 148. S. T. Grady, SATC, 8 0 Oak Ridge Turnpike, Oak Ridge, TN 37830 149. W. G. Greenman, Duratek Corporation, 8955 Guildford Road, Suite 280,
150. G. Gupta, Solar Thermal Electric Program, Solar Encrgy Research Institute,
151. J. 1E. Helt, Ph.D., Argonne National Laboratory, Director, 9700 South Cass
152. J. P. Hopper, Westinghouse Materials Co. of Ohio, P. 0. Box 398704,
153. R. L. Jacobson, Ph.D., University of Nevada Water Resources Center, Associate
154. M. W. Jugan, DOE, FOB, Rm. 2116, Oak Ridge, TN 37830 155. M. Lankford, 1J.S. DOE Headquarters, Trevion Bldg. 11, Suite 400,
156. P. B. Macedo, Duratek Corporation, 8955 Guildford Road, Suite 200,
157. T. L. Mayfield, FEDC, MS 8218, Oak Ridge, TN 37830 158. I,. W. McClure, Westinghouse Idaho Nuclear Company Inc,, P. 0. Box 4000,
159. '1%. McEvilly, 50E-111, Lawrence Berkeley Laboratory, 1 Cyclortron Road,
160. J. 0. Moore, DOE-OK, P. 0. Box 2001, Oak Ridge, TN 37831
Integration, LO-18, P. 0. Box 1970, Richland, WA 99352
Processing, P. 0. Box 464, Golden, CO 80402-0464
Iowa State IJnivcrsity, Ames, IA SO011
P. 0. Box 888, Richland, WA 99352
Juction, CO 81502-2567
MS417,Las Vegas, NV 89193-8521
Gcrniantown, MD 20874
Columbia, MD 21046
1617 Cole Boulevard, Golden, C(3 80401
Avenue - CMT/205, Argonne, IL 60439-4837
Cincinnati, OM 45329-8704
Research Professor, Suite 1 2505 Chandler Avenue, Las Vegas, NV 89120
12800 Middlebrook Road, Germantown, MS 20874
Columbia, M D 21046
Idaho Falls, ID 83403-4000
Berkeley, CA 94720
129
urphy, ET-AET, Eos Ala*los National Lab., NM 87545
Ridge Turnpike, Oak Ridge, Box 1410, C743 -TO& Paduca asurements Inc., P. 0.
ration, $955 Guildford Road, Suite 200,
ationa! Lab, MS FM3, h s Alamos,
, Oak Ridgc, 37830 . Box 628, Piketon, OH 45
178. J. L. Sleek, P.E., Manager, Waste Environ Remediation Programs, Westinghousc Savannah River Company, Box 616, Aikens, sc 29802
171. K. Stevenson, US. Department of Energy, St., New Yo& 172. @. Ik. Valle, Allied Signal Aerospace, Kan lant, P. 0. Box 41
173. P. ~~~~~~~~~ EG&G Idaho, Inc., V. 0. Box 1625, MS3940, Idaho Fall 174. J. Id. YOW, Jr., Law cnce Livermore National Laboratory, University
175. Ofliice of Assistant Manager, Energy Research and Development, DOE-OR,
176 - 185. Office of Scienlifk and Technical Information, P. 0. Box 62,
FV43, Kansas City, MO b4141-6259
1'- 0. Box 808, Livcrmure, CA 94550
1, Oak Ridgc, T N 37831
Oak Ridgc, TN 37831
. ..